LLM 관련 주요 논문 - 2026-07-17

1. Pretraining Data Can Be Poisoned through Computational Propaganda

Authors: Victoria Graf , Hannaneh Hajishirzi , Noah A. Smith , David Kohlbrenner , Kyle Lo
URL: https://arxiv.org/abs/2607.15267
Abstract:

Poisoning pretraining data can introduce harmful behaviors to LMs that are difficult to detect and mitigate. Prior work on poisoning pretraining data has largely exploited established data sources such as Wikipedia, which do not represent the large scale and heterogeneity typical of pretraining corpora, and has ignored the interaction between poisoned data and data curation pipelines. We demonstrate that poisoning attacks on pretraining data are feasible beyond this limited setting through an existing web-scale content injection mechanism: public discussion interfaces. Additionally, to measure whether malicious content is included after web crawling and data curation, we introduce HalfLife, a novel analysis for estimating adversarial content inclusion in web-crawl based LM training data. We use HalfLife to explore the feasibility of poisoning pretraining corpora at web scale through open discussion interfaces. Our analysis demonstrates the importance of estimating whether poison injections are included in pretraining data, and establishes third-party webpage content as a possible vector for attacking language model pretraining.

2. SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration

Authors: Yuyao Zhang , Junjie Gao , Zhengxian Wu , Jiaming Fan , Jin Zhang , Shihan Ma , Yao Yao , Weiran Qi , Chuyan Jin , Guiyu Ma , Xingzhong Xu , Kai Yang , Ji-Rong Wen , Zhicheng Dou
URL: https://arxiv.org/abs/2607.15257
Abstract:

Recent advances in Tool-Integrated Large Language Models have made web search a core capability of information-seeking agents. However, as interaction histories grow, agents increasingly struggle to track task progress. When search attempts fail to yield useful evidence, current single- and multi-agent systems can become trapped in repetitive loops, wasting search budgets and ultimately compromising the quality and completeness of the final output. We introduce SearchOS, a system-level multi-agent framework that turns fragile, implicit search progress into explicit, persistent, and shared state. First, we formulate open-domain information seeking as relational schema completion with grounded citations, where agents discover entities, populate attributes across linked tables, and anchor each value to source evidence. Then we design Search-Oriented Context Management (SOCM), which externalizes the evolving state into Frontier Task, an Evidence Graph, a Coverage Map, and Failure Memory. Built on SOCM, SearchOS applies a pipeline-parallel scheduling mechanism that overlaps the execution of sub-agents and continuously refills freed slots with tasks targeting unresolved coverage gaps to improve utilization and throughput. To schedule and control the execution of search agents, SearchOS introduces a Search Tool Middleware Harness that intercepts model and tool interactions to record grounded evidence and react to stalls or budget exhaustion, and provides a reusable hierarchical skill system comprising strategy and access skills to augment the agents’ search process and avoid repeating failed search patterns across runs. On WideSearch and GISA, SearchOS leads all metrics among the evaluated single- and multi-agent baselines, paving the way toward robust information-seeking collaboration.

3. teLLMe Why (Ain’t Nothing but a Jam): Exploratory Causal Analysis of Urban Driving Data

Authors: Qiwei Li , Jorge Ortiz
URL: https://arxiv.org/abs/2607.15254
Abstract:

Traffic agencies now have access to large volumes of video-derived data for studying safety and congestion. Most of these data are observational and collected without interventions, which makes causal questions such as “How would rain change traffic density?” difficult to answer. We present teLLMe, a system for exploratory causal analysis of urban driving datasets. The system starts from a structured event table built from dashcam annotations and combines causal structure learning with the PC algorithm, bootstrap-based stability checks, and query-specific effect estimation using linear regression and DoWhy. Natural-language questions are mapped to structured causal queries through a schema-aware LLM, enabling users to specify treatments, outcomes, and subpopulations. teLLMe returns a “Causal Card” that summarizes effect estimates, adjustment sets, DAG support, and assumptions, followed by a short natural-language explanation. Case studies on BDD-derived traffic events show that the system can surface plausible relationships involving weather, peak hours, and traffic density, while making uncertainty and modeling choices explicit. The system is designed as a tool for hypothesis generation and expert reasoning rather than a source of definitive causal claims.

4. When Words Are Safe But Actions Kill: Probing Physical Danger Beyond Text Safety in Hidden-State Risk Space

Authors: Weimeng Wang , Ziqiang Wang , Zihang Zhan , Chuanpu Fu , Qi Li , Ke Xu
URL: https://arxiv.org/abs/2607.15218
Abstract:

Large language models (LLMs) increasingly serve as high-level planners for embodied agents, where linguistically benign instructions can become unsafe once grounded in the physical world. We study whether this physically grounded danger is the same safety problem as ordinary text-level content danger. Through hidden-state direction analysis and random-split null tests, we show that content danger (CD) and physical danger (PD) form separable signals in LLM representations across Qwen2.5-3B/7B/14B/32B, Phi-3.5 and SmolLM2. Building on the CD/PD separability, we propose PRISM, a single-layer L2-regularized logistic probe over full hidden states. PRISM achieves 86.2–87.7\% accuracy on SafeAgentBench with 11.7–13.7\% FPR, while same-scale LLM judges over-block safe tasks at 24.7–39.0\% FPR. We further introduce PhysicalSafetyBench-1K (PSB-1K), a contrastive benchmark of 1{,}000 physical-risk pairs without direct harm keywords, to test whether methods detect physically grounded danger rather than explicit unsafe wording. On PSB-1K, PRISM reaches 99.6\% accuracy and 0.7\% FPR, whereas a Qwen2.5-3B judge rejects 67.8\% of safe tasks. PRISM also replicates on SafeText and EARBench, supporting hidden-state probing as a representation-level method for physical safety beyond text moderation.

5. Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation

Authors: Hoang-Loc Cao , Van Pham , Truong Thanh Hung Nguyen , Phuc Truong Loc Nguyen , Phuc Ho , Veronica Whitford , Hung Cao
URL: https://arxiv.org/abs/2607.15202
Abstract:

Annotation quality is a major bottleneck in building reliable and explainable artificial intelligence (XAI) systems for mental health research. In depression-related datasets, labels are often assigned without structured evidence, symptom-level justification, or traceable alignment with the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR), limiting both transparency and downstream model interpretability. We propose a self-evolving, expert-in-the-loop annotation framework for Major Depressive Disorder (MDD) that combines large language model (LLM)-assisted labeling with expert verification. The framework is intended to support the construction of explainable, DSM-5-TR-aligned datasets rather than to perform clinical diagnosis. It operates in three stages: candidate evidence selection from textual records, criterion-level DSM-5-TR analysis, and case-level synthesis that produces label-level diagnostic and severity annotations. A dual-memory architecture, composed of Example Memory and Reflection Memory, is designed to internalize expert feedback and iteratively improve future annotations without retraining. We describe this mechanism and leave its evaluation across multiple feedback cycles to future work. In addition to final labels, the framework exports clinical evidence, reasoning traces, and edit histories, enabling comprehensive auditability. In a pilot study using expert-reviewed samples, the proposed approach improves annotation consistency and explainability while reducing manual revision effort.

6. Can We Trust Item Response Theory for AI Evaluation?

Authors: Han Jiang , Sunbeom Kwon , Jinwen Luo , Ziang Xiao , Susu Zhang
URL: https://arxiv.org/abs/2607.15190
Abstract:

AI benchmarks increasingly leverage item-level statistical models, particularly item response theory (IRT), to estimate model capabilities, rank systems, select informative examples, and diagnose benchmark quality. However, AI benchmark data often departs from the data regime of human testing, for which standard IRT estimation tools were originally developed: benchmarks typically involve fewer evaluated models, far more items, and capability distributions that may be skewed, clustered, or multimodal. We examine how these regime mismatches challenge the reliability of IRT modeling for AI evaluation. Using item parameters and capability distributions derived from six widely used LLM benchmarks, we simulate response matrices under three common IRT models and compare four estimation tools used in recent benchmark studies: marginal maximum likelihood, Markov chain Monte Carlo, variational inference, and a neural pseudo-Siamese estimator. Across 18,000 simulation conditions, we systematically evaluate computational feasibility, scalability, and the reliability of IRT inferences about model rankings, predicted performance, and item characteristics. Results show that classical estimators can become infeasible in large benchmark settings, whereas scalable estimators can produce unreliable item-level and ranking inferences with small or nonnormally distributed model sets. This study identifies when latent trait models reliably support or risk distorting AI benchmarking claims, and what sample sizes and diagnostics are needed for trustworthy use.

7. Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy

Authors: Patrick Phuoc Do , Chau M. Ta , Chaoli Wang
URL: https://arxiv.org/abs/2607.15176
Abstract:

Multimodal large language models (MLLMs) are increasingly used to interpret visualizations, yet current evaluations remain largely chart-centric and provide limited evidence of understanding of scientific visualization (SciVis). We benchmark six MLLMs on the scientific visualization literacy assessment test, a standardized SciVis literacy assessment comprising 49 items based on 18 scientific visualizations and illustrations, spanning 8 techniques and 11 task types. We evaluate three closed-source and three open-source models under a closed-world protocol and compare their performance using data from 485 human participants. Results show that current MLLMs do not exhibit uniform SciVis literacy. Gemini is the strongest model overall, exceeding the human mean across the evaluated subsets, whereas the open-source models remain below the human baseline. Performance is highly uneven across techniques and tasks: models perform best on scientific illustration, search, and spatial understanding, but struggle on texture-based and integration-based visualizations and on quantitative estimation. Error analysis reveals recurring failures in fine-grained quantitative estimation, flow-direction interpretation, and grounded encoding interpretation. These findings position SciVis literacy as a necessary benchmark dimension for evaluating multimodal AI systems. Our code and model outputs are publicly available at this https URL .

8. Explaining Process Control Optimisation Recommendations via GradientSHAP and Implicit Differentiation

Authors: Paul Darm , Cem Alpturk , Kenneth Ulrich , William Duncan , Ali Anwar , Annalisa Riccardi
URL: https://arxiv.org/abs/2607.14970
Abstract:

Automated optimisation is increasingly adopted in industrial processes, yet a trust gap persists between engineers who design these algorithms and operators who must act on their recommendations. Explainable AI methods like SHAP (SHapley Additive exPlanations) have transformed interpretability for machine learning predictions; optimisation outputs could benefit from similar techniques. We present an approach that integrates Implicit Function Theorem (IFT) based sensitivity analysis with SHAP attribution and narrative generation via Large Language Models (LLM), producing explanations tailored for operators. Our approach leverages IFT to compute exact parameter sensitivities $\partial p^*/\partial x$ from the optimality conditions, enabling efficient GradientSHAP computation. For an industrial High Pressure Grinding Roll (HPGR) control optimisation problem with 22 features, we achieve equivalent SHAP attributions (correlation $>$0.99 with KernelSHAP) with over 40$\times$ speedup, enabling real-time natural language explanations. We validate on industrial scenarios and present feedback from domain experts on generated explanations.

9. Contextualized Early Detection of Online Firestorms: A Sequential LLM-Based Approach

Authors: Besim Shala , Peter Mandl , Andreas Humpe , Martin Häusl
URL: https://arxiv.org/abs/2607.14957
Abstract:

Online firestorms are rapid collective escalations of highly negative user-generated content and may cause substantial reputational and economic damage. Existing detectors usually work with volume signals, sentiment scores, or predefined linguistic features. Such signals are useful, but they capture contextual meaning shifts in evolving discussion threads only indirectly. This paper proposes an LLM-based detection system with two operating modes. The first mode classifies complete Reddit threads retrospectively by combining local chunk-level assessments into a thread-level judgment. The second mode processes threads sequentially and issues early warnings when a sliding window exceeds calibrated thresholds. In this mode, the language model estimates three firestorm indicators: negativity share, escalation level, and contributor count. On a balanced Reddit dataset, the global mode achieves strong classification performance, while the early warning mode reaches high recall and detects escalating threads after only a small number of comments and distinct contributors. The results indicate that LLMs can be used not only for static judgment tasks, but also as repeated estimators in context-aware monitoring of social media discourse.

10. Transcoders for Investigating Deception in Language Models

Authors: Darius Lim , Nathan Leow , Xin Wei Chia
URL: https://arxiv.org/abs/2607.14791
Abstract:

Transcoders have recently emerged as a promising approach for mechanistic interpretability (MI), enabling circuit-level analysis of model behaviour. In this paper, we investigate the use of transcoders to analyse deceptive behaviour in language models, a behaviour that poses a safety and security risk. Using a Qwen3-4B model with pre-trained transcoders, specifically per-layer transcoders (PLTs), we construct attribution graphs that capture feature activations and inter-feature dependencies, allowing circuit-level analysis of deception. Through feature steering and circuit analysis, we identified a dictionary of deception-related features and show that these features exert a stronger influence on deceptive outputs, as they produce predictable shifts between deceptive and non-deceptive responses. These findings suggest that deception emerges from internal model mechanisms and highlight the potential of transcoders for behavioural monitoring and early detection of security vulnerabilities related to malicious behaviours in language models.

11. AI vs Human Expert Reasoning: Assessing Agreements in Building Typology Predictions based on Street View Imagery

Authors: Zahratu Shabrina , Muhammad Asa , Jin Rui , Lu Yin , Stephen Law
URL: https://arxiv.org/abs/2607.14756
Abstract:

This research investigates the potential of Vision-Language Models (VLMs) to infer building typologies: Construction, Current Use, and Storeys from Google Street View (GSV) images. Predictions generated by VLMs are compared with inference by human experts (civil engineers and architects) as a source of manually labelled ground-truth data. We evaluate several state-of-the-art VLMs, including GPT-4o, Claude 3.5 Sonnet, and Gemini 2.0 Flash. By applying different scaling strategies and prompting techniques, we found that Chain-of-Thought prompts provide an overall more stable model performance. We also investigate the reasoning behind VLMs’ building-typology predictions by examining the probabilities of keywords appearing in AI explanations. This enabled us to analyse patterns in these reasonings and identify key themes driving both agreements and disagreements between VLM and expert labels. We find that AI tends to focus on visual indicators, whereas human experts place greater emphasis on broader contextual cues and domain knowledge, in addition to visual cues. Overall, VLM can approximate experts’ capability in building-typology classification at scale, with an average accuracy of approximately 70%. The study demonstrates the VLM’s potential for AI automation in tasks that require pattern recognition and object identification in an urban context. AI have the potential to serve as complementary and collaborative tools for urban analysis, leveraging their strengths in understanding visual patterns. This study contributes to the exploration of the efficiency and scalability of AI visual prediction and provides insights into the reasoning processes that could support automation processes in urban analysis and prediction.

12. Project Kaleidoscope: Contextual, Human-Aligned Evaluation for Real-World AI Applications

Authors: Leanne Tan , Rohan Jaggi , Shaun Khoo , Roy Ka-Wei Lee
URL: https://arxiv.org/abs/2607.14673
Abstract:

Evaluations (Evals) are a deployment bottleneck for real-world AI applications: public benchmarks rarely match a team’s users, context, or policies, and human review is often tedious to scale. Motivated by our work with AI applications in the public sector, this project addresses recurring evaluation challenges encountered when applications must satisfy local policy and governance requirements. We present Kaleidoscope, an integrated workflow for contextual functional evaluation that links persona-based test generation, contextualized rubrics, and human review for reliability-gated automated scoring. Generated test cases are scored against application-specific rubrics; human annotations provide reviewable labels; and LLM judges automate scoring only when their agreement with those labels meets a configured threshold. Kaleidoscope is therefore a practical, inspectable, iterative workflow for product teams. We report early evidence from a three-week pilot across four organizational use cases and custom-rubric judge experiments on 108 annotated Q\&A pairs spanning four domains and 14 evaluation dimensions. The results highlight useful features for end-to-end reliable, automated scoring.

13. SmartRAG: Native Graph-Based RAG for Mobile Device

Authors: Zhihan Jiang , Meng Li , Shenghao Liu , Keran Li , Ruiben Zhou , Xianjun Deng , Shuai Wang , Haipeng Dai
URL: https://arxiv.org/abs/2607.14661
Abstract:

Deploying large language models (LLMs) as personal assistants on mobile devices demands privacy, low latency, and offline availability, yet the computational cost of giant models clashes with strict edge-hardware budgets. We argue that this tension cannot be resolved by model compression alone; it requires decomposing on-device intelligence into complementary functional roles. We present SmartRAG, a fully on-device framework that organizes an intelligent assistant around four coordinated modules – Perception, Memory, Focus, and Thinking. At the core of SmartRAG is EvoNER, a continually learnable named-entity recognizer that incrementally expands its label inventory through teacher-distilled updates, enabling the system to absorb previously unseen entity types without retraining the backbone LLM. Extracted knowledge is stored in MRGraph, a three-layer provenance-preserving knowledge graph, and retrieved at query time through a hybrid pipeline combining graph traversal, lexical matching, and dense semantic search. The on-device LLM is invoked only for high-value semantic operations – labeling, planning, and answer synthesis – keeping inference costs bounded. Experiments on four QA benchmarks (TriviaQA, Natural Questions, HotpotQA, MultiHopQA) show that SmartRAG with a quantized 1.7B-parameter backbone achieves multi-hop reasoning performance competitive with models up to 18$\times$ larger, while running entirely on commodity smartphones within practical memory and latency envelopes.

14. TopoAgent: A Self-Evolving Topological Agent for Multimodal Scientific Reasoning

Authors: Mingze Xu , Yinghui Li , Jiayi Kuang , Zhanhui Kang , Di Yin , Ying Shen , Xing Sun , Yuxing Han
URL: https://arxiv.org/abs/2607.14658
Abstract:

While Multimodal Large Language Models (MLLMs) excel in general tasks, rigorous scientific reasoning remains challenging due to the limitations of monolithic, linear planning. Such sequential designs often suffer from visual-semantic misalignment, long-context hallucinations, and brittle execution under fixed task granularity. We propose TopoAgent, a self-evolving topological framework that replaces linear trajectories with dynamic, state-isolated graph evolution. TopoAgent first employs a front-end decomposer to fracture complex queries into visually-grounded atoms. These atoms are organized into a Directed Acyclic Graph (DAG) based on their dependencies, enabling strict context isolation to shield the reasoning engine from irrelevant historical noise. Furthermore, we introduce adaptive atomic fission, which dynamically splits bottleneck nodes into finer-grained sub-atoms at runtime when tool capability boundaries are exceeded. Extensive experiments across mathematics, physics, and chemistry benchmarks demonstrate that TopoAgent significantly outperforms state-of-the-art linear agent frameworks, providing a robust, noise-resistant, and self-correcting paradigm for autonomous scientific reasoning.

15. MCPEvol-Bench: Benchmarking LLM Agent Performance Across Dynamic Evolutions of MCP Servers

Authors: Huanxi Liu , Kun Hu , Jiaqi Liao , Qiang Wang , Pengfei Qian , YuanZhao Zhai , Dawei Feng , Bo Ding , Huaimin Wang
URL: https://arxiv.org/abs/2607.14642
Abstract:

As Model Context Protocol (MCP) servers emerge as the core infrastructure for connecting LLMs with external tools, existing benchmarks leverage real-world MCP servers to evaluate LLM agents’ tool-using capabilities. However, these benchmarks overlook the continuous evolution of tool interfaces and functionalities within MCP servers, resulting in flawed assessments that fail to capture the agent’s adaptability in changing tool landscapes. To bridge this gap, we introduce \textbf{MCPEvol-Bench}, a novel benchmark for evaluating the task-solving capabilities of LLM agents under dynamic toolset evolution. Inspired by large-scale empirical study, we propose 11 mutation operators to simulate realistic tool evolution within 123 MCP servers. We benchmark 12 state-of-the-art LLMs on multiple versions of MCP servers, revealing that even frontier models struggle to adapt to evolving tools. For instance, GPT-5.4 and Claude-Sonnet-4-6 exhibit performance declines of 13.7\% and 14.4\% in evolved MCP servers, respectively, accompanied by substantial increases in planning and reasoning errors. These findings highlight the vulnerability of LLM-driven workflows, establishing MCPEvol-Bench as a standard for evaluating agent adaptability in dynamic tool environments.

16. SportD: Can VLMs Physically Strategize?

Authors: Jasin Cekinmez , Addison J. Wu , Haotian Xia , Akshaya Bharadhwaj , Anay Putty , Anirudh Ravishankar , Jaewoong Lee , Jinglin Xiao , Kyumin Andrew Shim , Mishika Ahuja , Nisarga Patil , Leo Liu , Zhuohan Liu , Weining Shen
URL: https://arxiv.org/abs/2607.14616
Abstract:

Vision–language models have become increasingly capable of interpreting visual scenes, but it remains unclear whether they can use information to make strategically effective decisions. We investigate this question in soccer, where models observe the seconds preceding an on-ball decision and must choose whether to shoot or pass to a specific teammate. Unlike conventional visual-understanding tasks, soccer enables decisions to be evaluated quantitatively by estimating the value of every available action. We introduce SportD, a benchmark comprising 478 on-ball decisions from the 2022 FIFA World Cup. Each model choice is evaluated against a possession-value model that estimates the action that most increases the attacking team’s probability of scoring, allowing us to measure both optimal-action accuracy and the value forfeited by suboptimal decisions. Across three frontier VLMs, the best selects the highest-valued action on 31.4% of events, compared with 38.9% for the professional players, and all models incur significantly greater regret. Further analysis reveals a systematic preference for lower-variance and lower-reward actions: VLMs shoot less often and select substantially less progressive passes than either the optimal policy or the real players. The models also reproduce the player’s specific action above chance even when that action is suboptimal, suggesting partial imitation of familiar play patterns rather than consistent evaluation of counterfactual alternatives. SportD provides a value-grounded testbed for measuring physical strategic reasoning in VLMs.

17. MathCoPilot: An Interactive System for Human-AI Symbiotic Paradigm of Mathematical Research

Authors: Junjie Zhang , Jiayu Liu , Wenbin Liu , Zhenya Huang , Doudou Wang , Yan Jiang , Leiye Xu , Tao Xiong , Wen Huang , Qi Liu , Guoping Hu , Enhong Chen , Mengping Zhang , Xiangdong Ye
URL: https://arxiv.org/abs/2607.14582
Abstract:

Existing LLM-based theorem provers have achieved impressive results on formal mathematics benchmarks, yet they remain confined to acting as autonomous agents that prove a stated proposition. In this paper, we propose MathCoPilot, a human-in-the-loop system that embodies a new human–AI symbiotic paradigm for mathematical research, in which the mathematician steers the high-level mathematical direction while AI agents carry out the detailed formalization and proof work under continuous human guidance. MathCoPilot unifies three core capabilities: (1) an interactive workbench where the mathematician and AI agents collaborate through a living proof blueprint that decomposes a proof into navigable steps the human can directly inspect, direct, and refine; (2) automated proving skill orchestration with adaptive knowledge base search and Lean-integrated iterative verification; and (3) topic-driven paper retrieval and automated formalization into a verified Lean knowledge base. Using MathCoPilot, we systematically compare four state-of-the-art LLMs, including Gemini~3.1~Pro, GPT-5.4, and Claude~Opus~4.7, on a FormalMATH subset and on two real PDE theorems requiring deep domain expertise, evaluating their ability to produce verified Lean~4 proofs and to identify errors in deliberately incorrect proofs. Our results show that while current models can handle undergraduate-level problems with high success rates under favorable autoformalization conditions, substantial challenges remain for domain-specific theorems requiring genuine mathematical understanding.

18. Multi-LLM Collaborative MRI Report Generation for Visual Instruction Tuning in Brain Oncology

Authors: Sinyoung Ra , Jonghun Kim , Hyunjin Park
URL: https://arxiv.org/abs/2607.14581
Abstract:

Recent advances in large language models (LLMs) and their extension to vision-language models (VLMs) have made it easier to combine text and images for tasks such as report generation. Existing VLMs in medicine typically focus on 2D images (chest X-rays), and their extension to 3D imaging has been difficult because of the lack of paired 3D imaging-text data. Thus, we introduce a new method for creating a 3D image-text dataset for brain oncology using 3D MRI scans of glioma and meningioma cases. We use a cooperative system in which several LLMs work together to generate and check reports, ensuring that they are accurate and clear. By leveraging the new 3D MRI-text dataset, we further build a VLM that converts MRI scans into tokens and aligns them with text instructions. Our VLM performed better in report generation and visual question answering tasks than other 2D and 3D methods. Our method not only improves the quality of reports but also helps with better diagnosis and treatment in brain oncology.

Authors: Hao He , Chris J. Kuhlman , Xinwei Deng
URL: https://arxiv.org/abs/2607.14574
Abstract:

Collective problem solving often requires that group members consider the tradeoff between exploitation of known solutions and exploration for new ones, where information of known solutions can be disseminated among individual members through communication networks. The Mason–Watts experiment (PNAS 2012) showed that human groups in shorter-path networks outperform those in longer-path networks on a two-dimensional search task. In this work, we focus on the investigation of such network-efficiency effects in the setting of a group of large language model (LLM) agents. Specifically, we consider groups of sixteen LLM agents playing the Mason–Watts experiment on the eight Mason–Watts network topologies. Moreover, we develop mechanistic Bayesian optimization agents such that the performance of LLM agents can be compared with both the mechanistic agents and the human experimental data. Our computational experiments indicate that the LLM agents show a significant network-efficiency effect when instructed to randomize their first-round choices, but not under the default initialization. In this experiment, adding a one-sentence first-round randomization instruction improves collective payoff by more than three times the estimated payoff difference across the eight network topologies. Also, the Bayesian optimization agents obtain higher payoffs than the evaluated LLM agents on this spatial search task. We further compare the agents’ exploration–exploitation behavior, copying, and spatial diversity.

20. Alipay-PIBench: A Realistic Payment Integration Benchmark for Coding Agents

Authors: Shiyu Ying , Xuejie Cao , Yingfan Ma , Yuanhao Dong , Wenyu Chen , Bowen Song , Lin Zhu
URL: https://arxiv.org/abs/2607.14573
Abstract:

Payment integration is a demanding repository-level software task: agents must select a suitable product, implement coordinated client-server flows, verify payment outcomes, and preserve consistency between transaction and business states. We introduce Alipay-PIBench, a benchmark for evaluating coding agents on realistic Alipay payment integration. It contains nine product-specific projects and 18 task instances, each organized into Basic functional-completion and Advanced risk-aware hardening scenarios. Scenario-specific rubrics support deterministic static, unit, integration, and end-to-end checks, supplemented by LLM-assisted assessment for semantic requirements. We evaluate six coding-agent models and report rubric pass rate (RPR). Under the with-skill condition, mean RPR ranges from 68.58% to 91.37%. Access to the alipay-payment-integration skill improves mean RPR by 10.31 percentage points on average relative to the without-skill condition, with gains varying across models, products, and scenarios. Method-level results distinguish source-level completion, executable payment behavior, and payment-domain requirements. Alipay-PIBench provides a controlled setting for diagnosing model capability and evaluating structured guidance in payment integration.

21. Seeing the End at Step Zero: Accelerating Diffusion MLLMs via MLP Sparsity-Aware Truncation

Authors: Qicheng Zhao , Qi Sun , Zheyu Yan
URL: https://arxiv.org/abs/2607.14557
Abstract:

Diffusion Multimodal Large Language Models (DMLLMs) are highly effective for multimodal reasoning, yet their inference efficiency is significantly hindered by fixed-length generation constraints. Since the actual output length is unknown, output sequences are padded to a predefined maximum length, resulting in substantial redundant computation over unnecessary [EOS] tokens. In this work, we discover that DMLLMs implicitly reveal their valid semantic boundary at the very first denoising step through a distinct shift in MLP activation sparsity. Leveraging this observation, we propose Seer, a training-free framework that detects this boundary using a Signal-to-Noise Ratio (SNR)-based criterion and performs one-shot truncation of the redundant suffix for all subsequent computations. To preserve these theoretical gains during batched serving, Seer incorporates a hybrid execution strategy that maximizes throughput while seamlessly accommodating dynamic sequence lengths. Experimental results demonstrate that Seer effectively eliminates padding waste, accelerating throughput by up to $\sim$31$\times$. Across 9 benchmarks, Seer robustly maintains overall performance and even improves accuracy on complex visual tasks by mitigating noise leakage (e.g., DocVQA score increases from 63.52 to 63.66), offering a highly efficient, plug-and-play solution for DMLLM acceleration.

22. Towards an Intention Abstraction Layer for Autonomous Industrial Systems

Authors: Artan Markaj , Raphael Höfer , Felix Gehlhoff
URL: https://arxiv.org/abs/2607.14553
Abstract:

Modern industrial environments increasingly run many autonomous subsystems at once - schedulers, energy managers, vehicle fleets - each pursuing its own goals while sharing the same physical resources. Because high-level human intentions are translated into low-level control logic and then discarded, no running component can tell whether it is still doing what was actually intended, and goal conflicts surface only after they have caused a missed target or a shutdown. We propose the Intention Abstraction Layer (IAL), a domainagnostic middleware that represents intentions as first-class, persistent, and explainable runtime objects: a large language model grounded in a formal OWL ontology parses naturallanguage goals into structured intentions, a consistency monitor detects conflicts at registration time, before execution, and a transparency module explains them in natural language. We report a first proof of concept in which two autonomous agents register conflicting production and energy intentions, and the IAL flags and explains the conflict before it reaches the execution layer. The result is a mechanism that shifts behavioral assurance for cooperating autonomous systems from post-hoc failure analysis to pre-execution, intention-level checking.

23. Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization Agent

Authors: Lingyun Yang , Yuxiao Wang , Shenghao Liang , Linfeng Yang , Daocheng Ying , Chunbo You , Rui Zhang , Luping Wang , Yinghao Yu , Guodong Yang , Liping Zhang
URL: https://arxiv.org/abs/2607.14541
Abstract:

Existing GPU kernel generation benchmarks draw problems from synthetic or curated sources that diverge from deployed workloads. We present Atrex-Bench, a benchmark whose 30 operators and 440 shapes are sampled directly from full-cluster production inference traces of compute-limited, memory-rich GPUs. Each problem carries an importance weight derived from its share of observed GPU time, weighted by application card-hours and computed separately for the serving phases in which it runs, together with a per-problem roofline ceiling, so the aggregate score emphasizes the kernels that consume the most serving time. Evaluating six frontier coding agents on Atrex-Bench shows that even the best vanilla model reaches only ${\sim}10\%$ of the hardware roofline on production operators; and correctness alone overstates capability, since much of the apparent pass rate comes from PyTorch fallbacks rather than kernels the model wrote. To close this gap, we co-release Atrex-Kernel-Agent (AKA), a profile-driven kernel-optimization agent that combines iterative measure-revise search, optimization dropout for escaping stalled search contexts, and a layered GPU-optimization knowledge base (298 reference-kernel files and 244 optimization-knowledge documents, plus external upstream reference projects for API/ISA lookup). In a controlled case study, the agent converts zero-FlyDSL fallbacks into real kernels that match or exceed hand-tuned production baselines.

24. WrAFT: a Modularized Automated Writing Evaluation System for Argumentative Essays

Authors: Adnan Labib , Yixuan Huang , Jiahui Wu , John Maurice Gayed , Zheng Yuan , Qiao Wang
URL: https://arxiv.org/abs/2607.14524
Abstract:

This study presents WrAFT, a Writing Assessment and Feedback Tool, that delivers both accurate and reliable scores and effective comprehensive feedback to argumentative essays. WrAFT adopts a modular design by dividing automated writing evaluation (AWE) tasks into scoring, surface-level feedback, and deep-level feedback. In building the system, various Large Language Models (LLMs) have been evaluated, including LLaMA-3.3-70B-Instruct, GPT-4o, and Claude 3.7, through both direct prompting and supervised fine-tuning approaches. A proprietary dataset of 480 TOEFL Independent Writing essays with official benchmark scores was utilized. Benchmark-based evaluation shows that WrAFT achieves state-of-the-art performance in scoring, with a quadratic weighted kappa (QWK) of 0.84 and a root mean square error (RMSE) of 0.44 against official scores on a scale of 0-5. Human evaluation of system-generated feedback also reveals high approval ratings: 96.14 percent for surface-level feedback, 93.03 percent for deep-level macro feedback, and 94.69 percent for deep-level micro feedback. An interactive user interface has been developed for the system and is publicly available and free to use.

25. RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planning

Authors: Yanqiao Zhu , Jingru Gan , Xiaoqi Sun , Fang Sun , Yidan Shi , Md Mofijul Islam , Chao Shang , Wenhao Gao , Connor W. Coley , Yizhou Sun , Wei Wang
URL: https://arxiv.org/abs/2607.14512
Abstract:

Multi-step retrosynthesis planning seeks to decompose a target molecule into commercially available building blocks through a sequence of feasible reactions. The vast combinatorial search space makes this task challenging even for expert chemists. Traditional methods combine tree search with offline-trained value networks that score candidates in isolation, without reasoning about complete multi-step routes. Recent work leverages Large Language Models (LLMs) for this task, but relies on simple interfaces that limit exploration of the full search space. We introduce RetroAgent, an LLM agent that bridges symbolic search and neural reasoning through a harness with structured memory. Through memory and chemistry tools, the agent observes the full search state, including explored routes, available alternatives, and properties of intermediates, enabling informed decisions grounded in both global progress and domain knowledge. Experiments on in-distribution and out-of-distribution benchmarks demonstrate that RetroAgent delivers strong performance and generalization.

26. VLT: A Vision-Language-Time Series Multimodal Foundation Model for Industrial Intelligence

Authors: Haiteng Wang , Jingheng Yan , Xiaokang Wang , Lei Ren
URL: https://arxiv.org/abs/2607.14510
Abstract:

Industrial time series serve as the foundation for Prognostics and Health Management (PHM) to ensure the reliability and safety of industrial equipment such as aero-engines. However, existing approaches are typically limited to single-modality modeling, which restricts their generalization in complex scenarios. Although recent advances in large language models (LLMs) provide new opportunities for multimodal learning, bridging continuous time-series signals and discrete textual semantics remains an open challenge. To this end, we propose VLT, a multimodal foundation model that jointly models time-series, frequency-spectrum visual representations, and textual knowledge. A key insight is to utilize the frequency spectrum as a visual bridge to connect continuous temporal signals with discrete semantics. Specifically, a Time-aware Mixture-of-Experts (Time-MoE) is designed to capture heterogeneous temporal dynamics, while a Frequency-Text Augmented Learner enables joint modeling of spectral and semantic features within a shared representation space. Furthermore, a time-centric gradient alignment mechanism is introduced to mitigate cross-modal optimization conflicts via gradient normalization and reliability-aware dynamic reweighting. Extensive experiments on multiple industrial datasets demonstrate that VLT outperforms state-of-the-art methods, achieving superior robustness and generalization under few-shot, noisy, and incomplete-modality settings.

27. Contextualized Evaluation of Vision Language Models through Dynamic, Multi-turn Interactions

Authors: Yijiang Li , Huiqi Zou , Bingyang Wang , Ziang Xiao
URL: https://arxiv.org/abs/2607.14499
Abstract:

Multi-modal Large Language Models (MLLMs) have made substantial advances on benchmarks, yet their real-world effectiveness remains uncertain. This gap stems from the fundamental misalignment between benchmarks in controlled, static settings and the dynamic, interactive, and contextualized nature of real-world applications. To bridge this gap, we propose CEDI (Contextualized Evaluations of MLLMs through Dynamic, multi-round Interactions), a framework that recasts evaluation as a three-party interaction between an evaluatee model, an automated examiner, and a grader. The examiner conducts multi-turn, semi-structured conversation guided by a graph-based representation of the task. By navigating state-space transitions, CEDI deploys diverse strategies, from clarification requests to adversarial probes, to elicit performance evidence. We apply CEDI to visual hallucinations. Empirical results across multiple models, diverse settings, datasets, and domains show that contextualized, interactive evaluations reveal not only significantly more hallucinations than conventional static evaluation but also ones that more closely resemble those arising in practical use cases. We further show that hallucinations often accumulate over long contexts, through self-reinforcing dialogue history, and models are particularly vulnerable to questions requiring premise rejection or refusal. Together, these findings highlight CEDI as a step toward realistic, systematic, and ecologically valid assessments of MLLMs’ capabilities. Code is available at this http URL .

28. SAGA: Schema-Aware Grounding for Agentic Text-to-SPARQL Generation

Authors: Yiming Zhang , Koji Tsuda
URL: https://arxiv.org/abs/2607.14494
Abstract:

Complex knowledge base question answering (KBQA) is commonly approached through either information retrieval over a question-specific subgraph or semantic parsing into an executable logical form. We study the latter paradigm. Recent large language model agents make semantic parsing interactive: they alternate between reasoning, querying the knowledge base, and extending a partial SPARQL query. This interleaving reduces reliance on one-shot generation, but makes the quality of \emph{KB grounding} depend on what the interaction tools expose. Existing agents retrieve or prune candidate properties mainly through lexical relevance and instance-level observations, without systematically conditioning on entity types, property domains and ranges, or the expected answer type. We call this failure mode \emph{type-blind grounding}. It enlarges the grounding search space and often produces plausible-looking but semantically incompatible triple patterns that execute to empty results. We propose SAGA (\underline{S}chema-\underline{A}ware \underline{G}rounding for \underline{A}gentic Text-to-SPARQL Generation), a training-free framework that turns property exploration into a schema-constrained grounding operation. SAGA maintains a persistent bidirectional type state, filters known-incompatible property candidates at construction time, presents the remaining graph patterns in a compact schema-annotated format, and handles missing schema information permissively through empirical and trace-local evidence. Across nine benchmark settings over Wikidata and Freebase, SAGA achieves the highest F1 on all nine settings and the highest exact-match accuracy on eight, while reducing empty-result queries across all reported Wikidata settings.

Authors: Wenchang Gao , Pingyue Sheng , Lanlan Qiu , Yunfei Ma , Jian Zhao , Baicheng Chen , Kangda Wang , Yuyang Tian , Shunqiang Mao , Tianxing He
URL: https://arxiv.org/abs/2607.14485
Abstract:

Large language model (LLM)-based generative agents simulate human behavior through long-horizon decision-making processes that comprise intermediate steps such as planning, memory retrieval, reflection, and action selection. However, fine-grained human annotations of these intermediate steps remain scarce, and existing agents are not grounded in human preferences over such intermediate decisions. To address this gap, we introduce \method, an interactive simulation interface that enables us to collect step-level human preference supervision over agent decision trajectories, leading to a dataset of 57K fine-grained annotations. We conduct step-level preference learning on open-weight language models using supervised finetuning and direct preference optimization on this data, consistently improving simulation fidelity, coordination, and interaction quality, and inducing more socially effective agent behavior. Our results show that step-level human supervision is an effective training signal for improving both local decision quality and long-horizon agent behavior.

30. Reward-Free Evolving Agents via Pairwise Validator

Authors: Minghao Liu , Yu Wang , Jiayun Wang , Wei Wei
URL: https://arxiv.org/abs/2607.14408
Abstract:

A self-evolving agentic loop repeatedly proposes a tweaked version of an agent (its prompt template or program) and accepts or rejects the change based on a per-iteration quality signal. Designing that signal is often the costly part of the project: a reliable scalar reward requires domain expertise and labeled examples that are themselves as expensive to assemble as the agent’s underlying task. We propose replacing the scalar at the accept/reject gate with a pairwise validator: a frozen LLM that, given the parent and child candidate, returns a binary verdict on which is better. Pairwise judgment is generally easier and more stable than absolute scoring, due to its contrastive nature, which mitigates the need for strict scale calibration. The validator also requires no training of its own. We integrate the validator into three published self-evolving engines (GEPA, ADRS, ShinkaEvolve) and report two flavors: Adaptive Focus, which retains the engine’s existing val-set parent selection, and Soft Elo, which lets the validator’s verdicts drive parent selection so that val-set rewards drop as well. Across multiple agents and two artifact substrates (prompt and code), our method matches or exceeds the full-reward baseline on the majority of settings we evaluate, and the pattern survives a cross-family validator swap. The pairwise gate is thus a drop-in replacement for per-step reward design at competitive task accuracy without the labeling cost.

31. Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstration

Authors: Justin Bronder (Corabo Inc.)
URL: https://arxiv.org/abs/2607.14399
Abstract:

Evaluations of language-model honesty read the model’s verdicts as evidence about the model. We test the instrument instead. We built a text-adventure world where the game engine, not any model, knows whether the quest can be completed. A language model plays under a budget and must eventually declare its quest complete, unreachable, or not yet decidable; the engine scores every verdict. Decision rules were recorded before results were read, and run artifacts bind the revisions they executed; the strength of preregistration varies by series and is disclosed. With the player held fixed, instrument choices substantially changed measured behavior. On four byte-identical anchors, expanding a two-verdict grammar to three verdicts moved strong claims from 38/40 to 7/40, while the new incomplete verdict took 28/40 outcomes; across series 2, 93/158 valid games ended incomplete. One sentence disclosing the success criterion took matched-instance false verdicts from 18/59 to 0/58, through fewer decision points and cleaner decisions. Repeated runs of one fixed configuration produced non-stable verdict distributions on 3 of 4 instances: single runs report samples as dispositions. A formally preregistered narrative-register gradient was falsified; two post-hoc, hypothesis-generating patterns remain: register presence roughly doubled strong claims, and budget rendering moved verdicts more than register content (.383 meter vs .150 lantern). The narrator compressed abundant budgets toward scarcity landmarks, yet the registered mediation test returned a null. We propose a four-check integrity protocol for eval instruments.

32. CatalogAgent: A Supervisor-mediated Self-Learning System Enabling Context Engineering for GenAI Models

Authors: Zhu Cheng , Zhenming Wang , Yu (Xuan) Tang , Dan Liu , Bryan Zhang , Athanasios N. Nikolakopoulos , Pranav Souri Itabada , Jing Zhang , Chih-Chi Chou , Peng Gao , Fatemeh Mansoori , Bharat Bojja , Sarath Chander , Sameer Thombare , Umit Batur , Tarik Arici
URL: https://arxiv.org/abs/2607.14396
Abstract:

Product catalogs are the backbone of e-commerce sites, yet a large number of structured attributes (SAs) – such as material, color, and shape – often have missing values. Typically, SA values are extracted from product information, including titles and descriptions. While LLM-based generator-evaluator frameworks have demonstrated effectiveness for SA prediction – where an LLM generates SA values and another evaluates them – they face challenges when the Generator and Evaluator produce conflicting outputs, as either component can make mistakes. We introduce \texttt{CatalogAgent}, a novel agentic system that continuously improves Generator and Evaluator models for e-commerce catalog enrichment. When disagreements arise from (1) internal conflicts between the LLM-based Generator and Evaluator, or (2) external feedback from sellers on LLM outputs, a Supervisor Agent intervenes to mediate these conflicts and make final decisions. The system also incorporates a Memory Base and a Memory Summarizer that stores Supervisor Agent activities from individual cases and aggregates patterns into learnings. These learnings are fed back to the worker Generator and Evaluator LLMs, enabling self-improvement without human intervention. Through context engineering – injecting learnings and insights into worker LLMs’ contexts – the system successfully transfers the Supervisor’s capabilities to the Generator and Evaluator, improving their performance by 15.24\% and 13.98\%, respectively. Our experiments demonstrate a new paradigm of Supervisor Agent-mediated self-learning systems for improving generative AI model accuracy.

33. Chat2Scenic: An Iterative RAG-Based Framework for Scenario Generation in Autonomous Driving

Authors: Yuan Gao , Wenting Miao , Mattia Piccinini , Haoyu Wang , Qunying Song , Johannes Betz
URL: https://arxiv.org/abs/2607.14387
Abstract:

Validating autonomous driving systems requires diverse, regulation-compliant test scenarios. In simulation-based testing, scenarios are defined as executable scripts. Yet automatically generating such scripts from regulatory descriptions remains an open challenge, and existing approaches face fundamental trade-offs. Retrieval-assemble methods achieve reasonable compilation rates but lack scalability, whereas retrieval-based full-script generation suffers from low compilation success rates. We present Chat2Scenic, the first iterative retrieval-augmented framework to generate scenario scripts in Domain Specific Language (DSL). Specifically, Chat2Scenic provides a chatbot interface that supports interactive scenario refinement and integrates Retrieval-augmented Generation (RAG) to ground scenario generation in regulatory knowledge and DSL syntax. Furthermore, we propose an open benchmark for scenario generation comprising 123 scenarios from various regulations, including NHTSA and United Nations Vehicle Regulations, as well as other sources. Extensive evaluation with State-of-the-Art (SOTA) Large Language Models (LLMs) demonstrates that Chat2Scenic achieves 76.42% Compilation Success Rate (CSR) and 58.17% Framework Accuracy (FA), outperforming existing methods (Retrieval Assemble with 30.08% CSR, 11.03% FA and Retrieval full script generation with 16.26% CSR, 10.86% FA). To facilitate future research, we release our code as open source at this https URL .

34. CIPHER: A Decoupled Exploration-Selection Framework for Test-Time Scaling of Data Science Agents

Authors: Maxime Heuillet , Sharadind Peddiraju
URL: https://arxiv.org/abs/2607.14386
Abstract:

Data science tasks span from closed-ended information extraction to open-ended analysis, presenting significant challenges for automation. Recent AI agents powered by language models show promise for handling such complex tasks. However, existing agents typically rely on a single initial state that conditions the entire agent’s execution, making them vulnerable to cascading errors initiated by a suboptimal initial state. To mitigate this, we present CIPHER, an automated data science agent that leverages test-time scaling through the generation and selection of multiple initial states for concurrent execution. Unlike existing works on test-time scaling of AI agents, CIPHER explicitly decouples the generation of candidate initial states from their strategic selection for parallel execution. Through extensive evaluation on two benchmarks (closed-form and open-form tasks), we demonstrate that CIPHER exceeds state-of-the-art performance in matched-model comparisons, and remains competitive against larger-model baselines despite relying on a substantially smaller base LM. Our empirical study characterizes the design space of the Decoupled Exploration-Selection (DES) framework: we quantify how generation strategy, selection strategy, and aggregator model capacity contribute to overall performance, and derive actionable design recommendations for practitioners.

35. Traccia: An OpenTelemetry-Based Governance Platform for AI Systems

Authors: Nutan Kumar Naik , Aditya Kumar Saroj , Vijay Prasad Poudel , Saurav Samantray , Abhishek Patel
URL: https://arxiv.org/abs/2607.14309
Abstract:

The rapid development of Large Language Models (LLMs) and Artificial Intelligent (AI) powered autonomous agents has fundamentally changed the existing forms of software governance. In spite of the rigorous standards of transparency and account ability required according to the international frameworks such as the European Union’s AI Act, there is a considerable gap between theory and reality. The present study discusses the inherent drawbacks of currently utilized platforms for LLM evaluation, machine learning workflow, and application performance monitoring in general. It has been shown that current disjointed solutions fail to protect unbound state space agentic architecture from serious threats such as alignment drift, SaaS security concerns, and unauthorized deployment of shadow AI systems. Moreover, a solution is proposed for overcoming the discussed challenges in form of a coherent multi-level AI governance stack Traccia built on the top of OpenTelemetry infrastructure platform. Traccia resolves the last mile for AI Alignment by adding the telemetry data, passive semantic guardrail assessment, and execution lineage into a hashed trace ledger. Traccia automatically creates compliance evidence packages by appending tamper-resistant fingerprints and SHA-256 content hash, that map to regulatory requirements (Articles 12, 14, 19, 26(6), and 50 of the EU AI Act) without invading any data privacy. By performing this evaluation in a methodical manner, a solid machine-readable base has been created for enterprise-wide management of autonomous AI systems.

36. Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions

Authors: Zachary Izzo
URL: https://arxiv.org/abs/2607.14306
Abstract:

In this paper, we study the connection between an LLM’s output distribution and the data used to train it. Specifically, we study the degree to which an LLM’s next-token distribution agrees with the empirical next-token distribution (ENTD) given the context in the training data. The ENTD is an appealing target because it is the unrestricted global minimizer of the next-token cross entropy loss used for pretraining, as well as an easily interpretable function of the pretraining corpus. We find that for a significant fraction of inputs, the LLM’s distribution agrees with the ENTD almost perfectly, and the average agreement increases with model scale and training compute. Nevertheless, there is a long tail of input sequences where the LLM and ENTD differ significantly, and we examine several possible sources of this discrepancy across the transformer architecture, training procedure, and finite-sample noise in the ENTD estimate itself. More broadly, we hope our findings will encourage more work on ``data-centric mechanistic interpretability,’’ a complement to standard mechanistic interpretability that opens the black box of how model behaviors arise from the data, rather than how they are encoded in the learned weights.

37. AI Agents Do Not Fail Alone:The Context Fails First

Authors: Fouad Bousetouane
URL: https://arxiv.org/abs/2607.14275
Abstract:

Context engineering has become central to building reliable AI agents, yet it remains largely unmeasured. Agents do not fail in isolation: their behavior is shaped by the instructions, tools, memory, retrieved knowledge, guardrails, and untrusted inputs accumulated in their context. When this context is weak, agents drift, hallucinate, misuse tools, ignore constraints, become vulnerable to injection, and waste tokens. This paper validates context-engineering quality as an independent leading indicator of agent reliability. We implement the measurement in ProofAgent-Harness, an open-source infrastructure for AI agent evaluation that uses multi-juror, consensus-based scoring. The harness assesses context across seven criteria: role clarity, guardrail coverage, instruction consistency, tool schema quality, grounding sufficiency, injection hardening, and token efficiency. Crucially, the context score is isolated from behavioral metrics and release decisions, enabling a non-circular validation. Through a controlled context-quality study across regulated agent domains, holding frontier LLM agents fixed and varying only their operating context, we show that context-quality criteria consistently predict their corresponding behavioral outcomes. Grounding sufficiency predicts hallucination resistance, guardrail coverage predicts manipulation resistance, instruction consistency predicts instruction following, and tool-schema quality predicts tool use. These findings establish context measurement as a validated preflight signal for agent reliability and position context engineering as an auditable layer of agent evaluation and governance.

38. Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation

Authors: Genglin Liu , Muye Zhang , Krishnamurthy Viswanathan , Nichole J. Hansen , Blaž Bratanič , Nathan L Clement , Shalini Ghosh , Ariel Fuxman
URL: https://arxiv.org/abs/2607.14256
Abstract:

Multimodal Large Language Models (MLLMs) are increasingly deployed for nuanced content safety and moderation tasks, yet they remain vulnerable to adversarial attacks and out-of-distribution edge cases. Traditional active learning and manual annotation fail to scale against the complexity and volume of novel multimodal threats. In this paper, we propose an automated, agentic red-teaming framework that systematically synthesizes difficult examples using an iterative strategy that proposes novel hypotheses as well as mutating on past attempts. Leveraging a multi-agent architecture that consists of a high-reasoning Architect agent, an advanced image generator, and a multi-level verification committee of LLM raters, our system autonomously uncovers boundary-pushing violations and ambiguous policy edge cases without any human intervention. By employing these carefully synthesized adversarial examples as in-context demonstrations via test-time Retrieval, we substantially improve the target model’s robustness, reducing the False Negative Rate (FNR) from 41.2% to 24.5% in a public image safety benchmark without relying on any human labeling.

39. How Artificial Intelligence LLM Engines Shape the Global Conflict Information Environment

Authors: Jason Miklian
URL: https://arxiv.org/abs/2607.14197
Abstract:

Artificial Intelligence (AI) answer engines now field a growing share of the questions that analysts, scholars, and the public ask about issues of peace and conflict. Large Language Models (LLMs) are known to hallucinate under certain conditions, but do these errors have discernible patterns when they are asked about conflicts, and if so what can that teach us about the changing global conflict information environment? To answer, we first asked a battery of questions about 28 conflicts to five leading answer engines and scored their 5,460 answers against documented evidence. We found that the thinner the retrievable record around a given conflict, the more the engines invent, misattribute, and miscount. Thin records don’t just encourage hallucination, but create structural exposure to mis- and disinformation, because they are the easiest records to warp through Generative Engine Optimization (GEO) to bias engine responses. Through an analysis of 1,048 websites that the AI LLMs pulled conflict facts from, we found that GEO source optimization is already happening, and while state-partisan digital capture remains incipient it is rapidly growing. We explain what these findings mean for scholarship with the rise of GEO information warfare, and for policy argue for a return to the deep local monitoring and translation-based research that AI tools cannot replicate, closing with a discussion of future research opportunities and challenges in this fast-moving space.

40. RxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination

Authors: Haotian Liang , Mingkang Chen , Yufei Huang , Yuchun Guo , Xiaomeng Zhu , Xiangli Shi , Kaixuan Wang , Yunxuan Mao , Weijie Zhou , Ling Chen , Shirong Zeng , Yueyu Long , Yuchen Si , Yajuan Zhu , Xingyu Zhou , Minghui Wang , Wanjia He , Xin Yang , Lingzhu Xiang , Zhiqing Liu , Bohan Ma , Xiran Huang , Tianshuo Yang , Zhiheng Liu , Xuantang Xiong , Zisheng Lu , Ping Luo , Yao Mu , Han Hu , Zhengyou Zhang
URL: https://arxiv.org/abs/2607.14187
Abstract:

Embodied cognition requires agents to connect high-level task reasoning with the physical states to be achieved. We introduce Hy-Embodied-RxBrain, an embodied cognition foundation model with joint language-visual reasoning and imagination. Unlike vision-language models that emphasize scene understanding and textual decision making, or generative world models that mainly predict future visual states, RxBrain represents embodied plans in a single planning sequence where language and visual imagination play complementary roles. Language provides the abstract structure of a plan, including task decomposition, planning primitives, constraints, temporal order, and decision logic, while visual imagination grounds this structure through world state prediction and joint subgoal planning, associating each planning step with intermediate and final physical states. RxBrain adopts a unified multimodal Mixture-of-Transformers architecture that supports language, image, and video understanding and generation within one model. To train this capability, we build an automatic pipeline that converts embodied videos into joint text-visual planning supervision by decomposing videos into planning steps and aligning them with visual state transitions. We further introduce RxBrain-Bench to evaluate whether models can represent embodied plans through joint textual and visual components rather than separate understanding or generation. Experiments show that RxBrain maintains embodied understanding and generation abilities, and produces plans with coupled textual reasoning, world state prediction, and joint subgoal planning. We also extend RxBrain to continuous robot action generation, where it shows promising real-robot performance without large-scale action-data pretraining. These results provide an initial step toward foundation models for embodied cognition.

41. ReasFlow: Assisting Reasoning-Centric Scientific Discovery in Applied Mathematics via a Knowledge-Based Multi-Agent System

Authors: Yutong He , Daibo Li , Guohong Li , Jiahe Geng , Zhengyang Huang , Can Ren , Zekun Zhang , Yifan Liu , Shuchen Zhu , Hengrui Zhang , Boao Kong , Ming Sun , Shu Li , Chenyi Li , Jiang Hu , Kun Yuan , Zaiwen Wen , Pingwen Zhang
URL: https://arxiv.org/abs/2607.14178
Abstract:

Recent advances in Large Language Models have fueled autonomous AI agents capable of tackling complex scientific tasks, yet existing automated research systems remain predominantly focused on empirically driven domains with quantitative benchmarks, leaving theory-driven discovery, particularly in mathematically grounded disciplines requiring rigorous proofs and synthesis of domain knowledge, largely underexplored. Key challenges include the difficulty of verifying theoretical reasoning at scale, insufficient reasoning ability for autonomous frontier exploration, and a scarcity of procedural heuristics in the literature. We introduce ReasFlow, an end-to-end autonomous agent system for reasoning-centric scientific discovery that operationalizes a collaborative paradigm where the human expert acts as Principal Investigator while the agent executes rigorous derivations as a capable graduate student. ReasFlow incorporates (i) a robust internal verification loop that audits logical coherence and corrects fundamental errors prior to human inspection, and (ii) an automated knowledge retrieval and self-improvement mechanism that proactively surfaces both declarative facts and overlooked procedural heuristics, substantially reducing expert intervention. The system unifies literature synthesis, algorithm design, theorem proving, experimentation, and manuscript preparation in a single system. Deployed to autonomously generate five complete research papers with rigorous theoretical and empirical content from minimal prompts, ReasFlow consistently achieves the highest evaluation scores among state-of-the-art open-access baselines under a curated LLM-based review rubric. ReasFlow is publicly accessible via the ReasLab platform, providing a collaborative workspace for AI-assisted theoretical research. Github repo: this https URL .

42. When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models

Authors: Javier Aguilar Martín
URL: https://arxiv.org/abs/2607.14169
Abstract:

Large language models can synthesize a game’s rules as executable code - a Code World Model (CWM) - which a classical planner then searches over. Such models are typically accepted when they reach high transition accuracy on sampled trajectories. We argue this is the wrong notion of adequacy for planning. We show four things. (1) An LLM-synthesized CWM can pass a sampling gate at 100% transition accuracy and be $\geq 98\%$ state-accurate on the planner’s own search distribution, yet lose systematically at play, because the $<1\%$ it gets wrong is exactly the pivotal dynamics; the play cost of the omitted rule is $0.091$ (seed-clustered 95% CI $[0.065,0.117]$, $n=4800$). We call this the verified-vs-correct gap, and confirm it end-to-end through the synthesis pipeline. (2) The harm follows a quantitative law, $\mathrm{danger}=\mathrm{play_cost}\times(1-\mathrm{rarity})^N$, whose $(1-\mathrm{rarity})^N$ gate-miss factor is proven exact and whose play cost is empirically bounded. (3) The failure is not repaired by more data: LLM synthesis behaves as rule translation, not rule inference, and did not infer the omitted rule across models (GPT-5.x) and data regimes (including DAgger and targeted examples). (4) The same mechanism recurs on the belief-inference function of imperfect-information CWMs: we prove a coverage bound (a size-$N$ gate is identifying when $N\gtrsim b^{d_{\max} }$), explaining why shallow games such as Kuhn poker show no gap, and hand-construct Beacon, a verified-but-wrong inference function that passes the gate yet loses every game. These results suggest adequacy for planning-oriented world models should be measured on the search distribution or by play directly, not by prediction accuracy on sampled transitions.

43. MemoHarness: Agent Harnesses That Learn from Experience

Authors: Yue Huang , Wenjie Wang , Han Bao , Yuchen Ma , Xiaonan Luo , Yi Nian , Haomin Zhuang , Zheyuan Liu , Yue Zhao , Xiangliang Zhang
URL: https://arxiv.org/abs/2607.14159
Abstract:

An agent harness is the external control layer that turns a base LLM into an executable agent by managing context, tools, orchestration, memory, decoding, and output handling. While harness design strongly affects agent behavior, most automatic improvement methods optimize narrower artifacts such as prompts, pipelines, or workflows, and deployed agents usually reuse a single global harness for all cases. We introduce MemoHarness, an adaptive harness optimization framework that learns from its own executions. MemoHarness decomposes the harness into six editable control dimensions, stores per-case diagnoses and distilled global patterns in a dual-layer experience bank, and adapts the learned harness to each test case using retrieved experience without test-time labels, feedback, or additional search. In our evaluation across shell-agent, code-generation, and analytical-reasoning benchmarks, MemoHarness improves over the fixed harnesses we compare against and shows selective transfer to unseen suites and base models. Its additional context can also remain cost-competitive when much of the retrieved experience is cacheable. These results provide evidence that execution experience is a practical substrate for building agent harnesses that are more adaptive than a single static configuration, while leaving broader claims about statistical robustness and component attribution to future work.

44. Orchestrating Power Grid Studies with Multi-Agent AI and MCP Servers

Authors: Jérôme Picault , Clément Goubet
URL: https://arxiv.org/abs/2607.14158
Abstract:

This position paper explores how Agentic AI and Model Context Protocol (MCP) can support power-grid studies in a Transmission System Operator (TSO) context. We focus on integrating Large Language Models with numerical simulation tools, structured workflows, and human supervision. We identify key industrial requirements for agent assisted grid studies and introduce pypowsybl-mcp, an MCP-based interface exposing selected capabilities of our simulation tool, pypowsybl to AI agents. This first step provides a testbed to study how agents can setup simulations, execute analyses, retrieve results, and interact with power-system simulators through standardized tool calls. We also discuss principles for human-in-the-loop, multi-agent workflows and outline an evaluation strategy combining technical metrics and practitioner feedback. The paper positions MCP-based tool integration as a step toward more interactive, auditable, and scalable grid-study environments.

45. Enhancing Small Language Models Reasoning through Knowledge Graph Grounding

Authors: Dimitrios Kelesis , Konstantinos Bougiatiotis , Georgios Paliouras
URL: https://arxiv.org/abs/2607.14149
Abstract:

Although large language models (LLMs) have set benchmarks for zero-shot reasoning, their deployment remains cost-prohibitive and environmentally taxing. Small Language Models (SLMs) offer a sustainable alternative, but prone to errors, on tasks requiring complex, multi-hop logical grounding. We investigate a neuro-symbolic agentic framework to enhance the reasoning capabilities of SLMs, specifically Gemma 3 (1B, 4B) and Llama 3.2 (3B), using the CLUTRR kinship benchmark. Our approach transforms the SLM into a minimalist agent utilizing two specialized tool calls: extract_facts for symbolic triplet extraction and get_hint for expert reasoning via a Relational Graph Convolutional Network (RGCN). We evaluate these models across two configurations, both in an Oracle scenario with ground-truth triplets and a Realistic scenario relying on self-extracted knowledge. Our results reveal that while RGCN-derived hints provide a 1.5 - 2x performance gain over story-only baselines, the system is constrained by the extraction bottleneck and sequential deductive fragility, where early extraction errors compound over multi-hop chains. Furthermore, we identify a “distraction effect” in specific architectures where noisy, self-generated facts degrade performance despite the presence of expert hints. This work characterizes the challenges of symbolic grounding in low-resource agentic systems and provides a roadmap for iterative verification in neuro-symbolic agentic pipelines.

46. ToolAnchor: Anchoring Counterfactual Context to Boost Agentic Tool-use Capability

Authors: Weiting Liu , Jieyi Bi , Wanqi Zhou , Jianfeng Feng , Yining Ma , Ai Han , Wenlian Lu
URL: https://arxiv.org/abs/2607.14145
Abstract:

Tool-augmented large language model agents excel at long-horizon tasks, yet they are typically post-trained on fixed toolsets. When tasks demand new tools, these agents struggle to incorporate them effectively, and retraining from scratch is often impractical. We identify the core obstacle in such toolset expansion problem as behavioral inertia: the tendency of agents to fall back on familiar tools and established reasoning patterns despite having access to new ones. We demonstrate that injecting counterfactual anchor contexts at critical decision points can break this inertia, recovering failed trajectories by eliciting suppressed agent capabilities. To scale this insight, we propose ToolAnchor, a framework that uses teacher models to hypothesize these counterfactual contexts, verifies them via student rollouts, and internalizes the successful interventions through agentic post-training. Extensive evaluations across general AI assistant (GAIA), textual search (BrowseComp), and visual search (VDR-Bench) tasks demonstrate that ToolAnchor consistently exhibits competitive performance under expanded toolsets. Our work bridges the gap between static post-training and dynamic adaptation, charting a new path for scalable agentic reinforcement learning.

47. Human AI Construction of Bayesian Networks for Operational Decision Support – A Virtual Survey Approach

Authors: Kumar Rahul (1), Shovan Chowdhury (2) ((1) Indian Institute of Management Kozhikode, Kerala, India, (2) Indian Institute of Management Kozhikode, Kerala, India)
URL: https://arxiv.org/abs/2607.14141
Abstract:

Bayesian Belief Networks (BBNs) are powerful tools for decision-making under uncertainty. However, building their structures and estimating parameters are difficult. Currently, researchers must choose between relying on expert judgement or using large datasets to learn the structure and parameters of the network. We propose a new methodology using Large Language Models to bridge the gap between expert opinion and data-driven learning. This approach uses a panel of AI agents to estimate probabilities based on specific personas and context. We then apply a trimmed-mean rule to remove noise from these responses. We develop a six step BBN framework and illustrate it to model customer intention to consult a doctor in an alternative healthcare system. The model reveals that while self efficacy appears to be a major factor, its actual causal impact is small. In contrast, subjective norms have a much stronger effect in modelling customers’ intention. The most effective strategy is to improve both confidence and community norms simultaneously.

48. Interpretable Language Model for Closed-Loop Type 1 Diabetes Control

Authors: Maya Sarkar
URL: https://arxiv.org/abs/2607.14126
Abstract:

Type 1 Diabetes (T1D) is a chronic, life-threatening autoimmune condition characterized by the complete destruction of insulin-producing pancreatic beta cells. While Artificial Pancreas Systems (APS) powered by Reinforcement Learning (RL) have shown promise in automating insulin delivery, their ``black-box’’ nature makes it hard for patients and doctors to trust them fully. This paper presents LLM-T1D, a promising approach that combines the precision of RL with the clear, human-like reasoning of Large Language Models (LLMs) to create a more transparent and reliable insulin pump controller. By training an expert RL system and distilling its knowledge into fine-tuned LLaMA 3.1 8B and Qwen3 8B models, we developed a controller that not only surpasses the RL system’s performance but also explains its decisions in plain, understandable language. Tested on the FDA-approved UVA/Padova T1D simulator, the LLM controllers deliver excellent blood sugar control (73.5% Time in Range) while maintaining strict formal safety verification against hallucinations.

49. HG-RAG: Hierarchy-Guided Retrieval-Augmented Generation for Structured Knowledge Graphs

Authors: Pranav Yadav
URL: https://arxiv.org/abs/2607.14095
Abstract:

Retrieval Augmented Generation (RAG) has proven to be a widely successful process at improving the quality of outputs from a Large Language Model (LLM) for wider context. However, RAG systems typically retrieve context from flat document stores, which struggles when queries require hierarchical or relational reasoning across structured knowledge. I present HG-RAG (Hierarchy-Guided RAG), a framework that performs graph-traversal over a hierarchical knowledge graph to deliver structured context to a language model. My retrieval pipeline resolves a named entity anchor from the query, then expands context upward through parent nodes, laterally through relational neighbors, and downward through child nodes when needed. I evaluate HG-RAG against a dense retrieval baseline across three world scales (18-800 nodes) with four query types: local fact, hierarchical, neighborhood, and multi-hop. Results show HG-RAG consistently outperforms the flat baseline on hierarchical, relational, and multi-hop reasoning tasks, while reducing hallucination and maintaining locality coherence.

50. In-Place Tokenizer Expansion for Pre-trained LLMs

Authors: Jimmy T.H. Smith , Tarek Dakhran , Alberto Cabrera , Simon S. Lee , Paul Pak , Aditya Tadimeti , Tim Seyde , Maxime Labonne , Alexander Amini , Mathias Lechner
URL: https://arxiv.org/abs/2607.15232
Abstract:

A tokenizer fixed at the start of pre-training allocates vocabulary in proportion to the pre-training corpus, reflecting the deployment priorities at that time. When those priorities shift, languages added later are split into many more tokens per word, which can raise latency, compute, and energy consumption for users of those languages. Cloud models can afford a broad vocabulary because the embedding and LM-head matrices are a small fraction of their parameters. On a compact model those matrices are a material share of per-token decode bandwidth, so on-device models ship small vocabularies and accept fragmentation outside a fixed language set. We present tokenizer expansion, an in-place recipe for upgrading a pre-trained model’s tokenizer when the model producer controls its design. We continue the existing tokenizer’s BPE merges on a multilingual corpus, so most source tokens carry over unchanged as single tokens and every new token has an exact decomposition into source tokens. We copy the carried-over embedding rows unchanged and initialize new rows as the mean of their source sub-token embeddings. A two-stage adaptation, embedding-only training then full-model continued pre-training, recovers source-checkpoint quality. We apply the recipe to a continued pre-trained checkpoint of LFM2-8B-A1B, an 8B-parameter Mixture-of-Experts model, to help produce LFM2.5-8B-A1B with a 128K tokenizer. The expanded tokenizer encodes Hindi and Vietnamese in roughly $2.4\times$ and $2.6\times$ fewer tokens than the source (up to $4.0\times$ on Thai). Combining these reductions with the measured per-token cost of the larger vocabulary, we estimate a $2.2$-$3.7\times$ per-character decode speedup for these languages across our reference devices. We release the model weights and the expanded tokenizer, and report the negative findings that shaped the recipe.

51. Symbal: Detecting Systematic Misalignments in Model-Generated Captions

Authors: Maya Varma , Jean-Benoit Delbrouck , Sophie Ostmeier , Akshay Chaudhari , Curtis Langlotz
URL: https://arxiv.org/abs/2607.15216
Abstract:

Multimodal large language models (MLLMs) often introduce errors when generating image captions, resulting in misaligned image-text pairs. Our work focuses on a class of captioning errors that we refer to as systematic misalignments, where a recurring error in MLLM-generated captions is closely associated with the presence of a specific visual feature in the paired image. Given a vision-language dataset with MLLM-generated captions, our aim in this work is to detect such errors, a task we refer to as systematic misalignment detection. As our first key contribution, we present Symbal, which utilizes a structured, dual-stage setup with off-the-shelf foundation models to identify systematic misalignments and summarize results in natural language. As our second key contribution, we introduce SymbalBench, a benchmark designed to evaluate automated methods on our proposed task. SymbalBench consists of 1.7 million image-text pairs from two domains (natural and medical images), organized into 420 vision-language datasets with annotated systematic misalignments. Symbal exhibits strong performance on this benchmark, correctly identifying systematic misalignments in 63.8% of datasets, a nearly 4x improvement over the closest baseline. We supplement our evaluations on SymbalBench with real-world evaluations, showing that (1) Symbal can accurately surface systematic misalignments in captions generated by four MLLMs and (2) Symbal is a powerful tool for auditing off-the-shelf image-caption datasets. Ultimately, our novel task, method, and benchmark can aid users with auditing MLLM-generated captions and identifying critical errors, without requiring access to the underlying MLLM. Code is available at this https URL .

52. MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization

Authors: Shaoxiong Zhan , Shi Hu , Boyu Feng , Hai Lin , Andrew Gong , Zhengda Zhou , Jiaying Zhou , Yunyun Hou , Hao Su , Hai-Tao Zheng
URL: https://arxiv.org/abs/2607.15205
Abstract:

Real repository issues routinely include visual evidence such as screenshots, error dialogs, rendered UI states, and logs, yet repository-level issue localization is evaluated mostly as a text-only task. Existing multimodal SE benchmarks evaluate end-to-end repair, entangling localization with patch synthesis and obscuring whether visual input helped, hurt, or was ignored. We introduce \textbf{MM-IssueLoc}, a controlled benchmark and evaluation protocol for repository-level localization with visual evidence. MM-IssueLoc contains 652 issue-PR instances across 23 languages, with annotations for 7 image categories and 4 relevance levels. It provides file-level and function-level gold labels, paired text-only and with-image evaluation, and VCE-based diagnostics that convert images into structured textual evidence. We evaluate LLM-based and retrieval-based systems, including MM-IssueLoc-VL-Emb as a controlled multimodal retriever. Results show that existing systems remain far from reliable multimodal repository localization: the strongest agent reaches 38.96 file Acc@5 and 22.45 function Acc@10, while the strongest retriever reaches 33.86 function Acc@10. Cross-benchmark comparisons show that high localization scores on text-dominant SWE benchmarks do not transfer cleanly to multimodal issue localization. MM-IssueLoc turns visual evidence into an explicit evaluation variable, enabling future work to test whether systems improve by using visual evidence for localization, rather than by relying on text-only cues or downstream patch-generation effects.

53. Mask-Aware Policy Gradients for Diffusion Language Models

Authors: Haran Raajesh , Kulin Shah , Adam Klivans , Philipp Krähenbühl
URL: https://arxiv.org/abs/2607.15200
Abstract:

Reinforcement learning has proven effective for improving reasoning in large language models, but extending it to Masked Diffusion Language Models (MDLMs) remains challenging due to the intractability of the log-likelihood estimation. Existing approaches approximate this log-likelihood by modeling only the token predictions, ignoring the order in which positions are unmasked during generation. We observe that MDLM generation involves two decisions at each step: what tokens to place at each masked position and which positions to remask. We formalize this as a two-stage action MDP, showing that the policy gradient naturally decomposes into a token term and a masking term. Combining optimization of both terms leads to state-of-the-art outcomes on mathematical reasoning and coding benchmarks, with scores of 87.1% on GSM8K and 53.4% on MBPP.

54. Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents

Authors: Dylan Van Mulders , Matthias Bogaert , Dirk Van den Poel
URL: https://arxiv.org/abs/2607.15095
Abstract:

The formation of political coalitions is a complex negotiation driven by both concrete policy objectives and deep-seated ideological convictions. While Large Language Models (LLMs) open new avenues for computational political science, the neutrality and helpfulness biases instilled by Reinforcement Learning from Human Feedback (RLHF) prevent them from sustaining steadfast partisan behaviour. We present a multi-agent framework that reconciles factual grounding with ideological alignment by combining Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Retrieval-Augmented Generation (RAG): DPO instils aggressive party-specific personas, while a per-party RAG pipeline keeps each agent bounded to its official manifesto. We operationalize the framework on the 2019 Flemish election, deploying the partisan agents in a hub-and-spoke negotiation arbitrated by a formateur. To make the emergent negotiation interpretable, we introduce a Multi-Layered Information Lineage Topology (MILT) that traces every clause in the final agreement back to its manifesto origin and classifies it into five provenance states, a Coalition Influence Score (CIS) that aggregates these traceable contributions to identify which party shaped the agreement, and a real-world grounding pass that benchmarks each simulated provision against the historically adopted coalition agreement. Across three independent simulations the framework yields a stable winner and ranking (N-VA ahead of CD\&V and Open Vld), and manifesto-anchored lineage reliably predicts real-world materialization whereas hallucinated content does not. The result is a transparent, scalable testbed for the ex-ante exploration of party compatibility and formateur-mediated compromise.

55. OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios

Authors: Chengyu Shen , Yujie Fu , Gangtao Xin , Yanheng Hou , Wenlong Fei , Guojie Zhu , Jiawei Li , Hongcheng Gao , Runming He , Zhen Hao Wong , Meiyi Qiang , Hao Liang , Zhao Cao , Hao Jiang , Chong Chen , Wentao Zhang
URL: https://arxiv.org/abs/2607.14989
Abstract:

Large language models are increasingly evolving from text generators into general agents capable of understanding user requests, invoking external tools, and completing complex tasks through interaction. However, existing agent benchmarks often focus on limited scenarios, tool ecosystems, or interaction formats, making it difficult to systematically characterize model capabilities across heterogeneous application settings. We introduce OmniaBench, a benchmark for evaluating general agents across diverse scenarios with explicit state spaces. We derive application-oriented scenario knowledge from app stores, product documents, industry resources, Web retrieval, and human refinement, forming a hierarchical taxonomy that spans ToC, ToB and ToE with 90 level-1 and 354 level-2 domains. Based on this taxonomy, we construct executable environments and synthesize single-turn and multi-turn tasks through four complementary routes: DAG, DAG-S, Solver, and Program. OmniaBench further introduces a ten-dimensional capability taxonomy and eight compositional atomic difficulty factors to support fine-grained evaluation and analysis. The resulting dataset contains 1,431 tasks, together with a challenging subset of 644 tasks designed to reduce evaluation cost and mitigate potential contamination of the full set after public release. The bench presents substantial challenges to current frontier models, with even Claude-Sonnet-5 and GPT-5.6-Sol achieving Overall Pass@1 scores of only 58.54 and 57.14, respectively. Further analyses reveal clear differences across domains and capabilities, as well as persistent limitations in planning, constraint maintenance, and adaptive correction. OmniaBench provides a broad and diagnostic benchmark for characterizing the capability boundaries of general agents.

56. Show Me How You Reason and I’ll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution

Authors: Zlata Kikteva , Artur Romazanov , Annette Hautli-Janisz , Ramon Ruiz-Dolz
URL: https://arxiv.org/abs/2607.14905
Abstract:

Given the current trend to employ large language models (LLMs) in almost any imaginable context, LLM-generated text detection and authorship attribution have become a pressing issue. Prior work has primarily focused on surface-level linguistic features, an approach shown to be susceptible to paraphrasing and other obfuscation techniques. In this paper, we go beyond the linguistic surface, extracting and analysing reasoning structures in LLM-generated texts with the goal of capturing more complex signals of LLM authorship. We propose a graph neural network approach that leverages reasoning graphs extracted by an argument mining pipeline, demonstrating improved robustness and generalisation over a traditional Longformer baseline. Our approach outperforms the baseline by up to 27 percentage points under the obfuscation attacks such as paraphrasing and backtranslation, and 19 percentage points when evaluated on the texts generated by the unseen model versions, simulating real-world conditions in which new LLM versions are continuously released.

57. StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows

Authors: Sizhong Qin , Yi Gu , Yao Jiang , Ao Cai , Changjian Zhou , Shaoxuan Shuai , Jiachang Wang , Tianhao Shen , Yueqiang Li , Xinhao Li , Li Zeng , Yueshi Chen , Dachen Gao , Genrong Xu , Wenjie Liao , Xinzheng Lu
URL: https://arxiv.org/abs/2607.14896
Abstract:

Addressing a structural-engineering request requires more than a single answer; it requires a chain of interdependent artifacts: interpreted requirements, a computable model, validation records, solver outputs, code-check records, and a final report. Evaluations centered on question answering or script generation rarely verify this complete evidence chain and may therefore reward fluent outputs even when the underlying engineering workflow is incomplete, internally inconsistent, or non-executable. To address this limitation, we present StructureClaw, an artifact-centered workbench in which LLM agents operate through governed engineering skills, typed tools, shared artifact state, and local analysis backends. We also introduce StructureClaw-Bench, an executable benchmark of 150 controlled scenarios spanning standard workflow execution, interactive robustness, and multimodal structural-model reconstruction. A scenario succeeds only when all required artifact- and execution-level assertions pass in a single run. Across ten agent-model configurations, each evaluated on the same 50 standard cases, the average Success Rate rises from 56.8% with the generic-skill baseline to 88.6% with the full automatic workflow. The interactive and multimodal evaluations identify two prominent remaining challenges: safe handling of invalid numerical inputs and fixture-consistent reconstruction of structural models. These findings show that artifact-centered evaluation can expose workflow-level failures that are difficult to identify from final responses alone, providing a more rigorous basis for evaluating and improving structural-engineering agents. The code and benchmark are available at this https URL .

58. Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs

Authors: Robert Graham , Edward Stevinson , Yariv Barsheshat
URL: https://arxiv.org/abs/2607.14888
Abstract:

Finetuning language models on small, curated datasets is standard practice for adapting them to specific policies or domains. We show that finetuning on narrow, factually-defensible, moderation-passing data can cause broad ideological shifts across unrelated domains, while preserving general capabilities. Training GPT-4.1 on right- or left-leaning economics Q&A yields matched ideological shifts on topics such as criminal justice, the environment, and cultural taste. The same effect appears with plausibly-deployed datasets such as workplace HR policy and practical finance queries, as well as on a science-pseudoscience axis where food-safety finetuning increases sycophantic agreement with users expressing false health beliefs. We call this phenomenon ideological generalisation and propose a methodology to measure two properties: breadth, how far the shift reaches across topics absent from training, and amplification, how much finetuning intensifies the shift relative to few-shot prompting on the same examples. We show that few-shot prompting indicates the direction of generalisation but finetuning pushes the model to further extremes, including to far out-of-distribution outputs such as endorsements of race-IQ connections and political violence. The effect replicates on Gemma-3, holds under judge-free evaluations and external benchmarks, survives mixing with generic data, and leaves GSM8K accuracy within $\pm 1$pp of the baseline.

59. Does generative AI supersede supervised XMLC? A Benchmark Study on Automated Subject Indexing with German Scientific Literature

Authors: Maximilian Kähler , Katja Konermann , Lisa Kluge , Markus Schumacher
URL: https://arxiv.org/abs/2607.14882
Abstract:

With a large controlled vocabulary as the label set, the task of automated subject indexing in a library can be understood as a multi-label classification task. If the set of subject terms is large, the problem fits the Extreme Multi-Label Classification (XMLC) objective. In this study, we apply a selection of specialised supervised XMLC methods to the test case of subject indexing contemporary German scientific literature, collected at the German National Library (DNB). We contrast these results by including a classical lexical matching baseline and three of our own recently developed LLM-based methods into the benchmark. Algorithms are evaluated and compared in several metrics. This includes binary relevance comparisons with previously indexed material, as well as graded relevance ratings by professional subject librarians. A challenge for all methods is to reliably make suggestions from the long tail of the subject vocabulary. We find that supervised XMLC algorithms relying on transformer-based dense features give best results in terms of overall binary relevance metrics. However, focusing on graded relevance and performance in the long tail of our subject vocabulary, the LLM-based generative methods give better results, making them a promising alternative for future productive use.

60. Can LLMs Build a MaxSAT Solver from Papers? The CoreForge Experience

Authors: Ruben Martins
URL: https://arxiv.org/abs/2607.14818
Abstract:

We report on CoreForge, an experience in using large language models (LLMs) to build an unweighted MaxSAT solver from research papers rather than from an existing solver codebase. The project focuses on unsatisfiability-based MaxSAT algorithms and follows an iterative workflow that combines paper discussions with ChatGPT, implementation through Codex prompts, and repeated LLM-assisted code audits and revisions. Although the codebase implements several algorithms and solver components, our evaluation focuses on configurations that combine core-guided optimization, lightweight preprocessing, core minimization, integration with integer linear optimization backends, and a new core-sequence lookahead approach. Our experience suggests that LLMs can support solver implementation from papers, while requiring external validation, benchmarking, and human guidance. In our experiments, fuzzing and MaxSAT Evaluation instances did not reveal wrong answers in the tested configurations, although performance remains below the best hand-engineered MaxSAT solvers. We summarize what worked, what remained difficult, and the lessons for future LLM-assisted solver development.

61. Large Language Models for Code Generation from Multilingual Prompts: A Curated Benchmark and a Study on Code Quality

Authors: Saima Afrin , Alessandro Midolo , Camilo Escobar-Velásquez , Mario Linares-Vásquez , Weiyuan Ding , Bowen Xu , Massimiliano Di Penta , Antonio Mastropaolo
URL: https://arxiv.org/abs/2607.14816
Abstract:

Large Language Models (LLMs) perform differently on identical programming tasks when prompted in different natural languages, a phenomenon known as language bias. While this behavior has been widely studied for general text generation, its impact on code generation quality and programming conventions remains largely unexplored. We investigate how the language used to describe programming tasks affects the source code generated by GPT-4o mini, DeepSeek, and Claude. Our study comprises 460 coding tasks spanning Python (230) and Java (230). We translate and manually curate the original English prompts into Chinese, Hindi, Spanish, and Italian while preserving their technical meaning. We evaluate the generated code using multiple dimensions, including functional correctness through test pass rates, structural quality using established code metrics, issues detected by static analysis tools, and lexical characteristics such as the language used in identifiers and comments. Our results show that (i) English prompts do not consistently produce the best functional correctness or code quality, (ii) the impact of prompt language depends on both the programming language and the LLM, and (iii) generated code frequently mixes English with the prompt language in comments and string literals. These findings provide the first curated multilingual benchmark for studying language bias in code generation and offer insights for developing more robust multilingual code generation systems.

62. Dialogue Summarization with Emotion Dynamics Using Topic- and Participant-Centric Decomposition

Authors: Linyun Xiang , Mark Neerincx , Stephanie Tan
URL: https://arxiv.org/abs/2607.14769
Abstract:

Existing text summarization research has focused much on monologic information (e.g., newspaper articles, reports) without accounting for the interaction between speakers or authors. In contrast, dialogues are a rich communication channel where multiple participants conduct back and forth exchanges to construct meaning. We propose a dialogue summarization framework that explicitly models both semantic and emotion dynamics using multimodal dialogue inputs, built on an adapted hierarchical Chain-of-Agents approach. We decompose dialogues from two perspectives: (1) topic segments based on the utterances of all participants, and (2) participant-specific utterance segments. These are used to generate corresponding summaries while incorporating automatically inferred emotions. Topic- and participant-level summaries are aggregated into a dialogue summary capturing semantic content and emotion trajectories. To evaluate beyond content accuracy, we introduce emotion trajectory metrics measuring how well summaries preserve emotional flow. Experiments with small language models on multimodal dialogue datasets show that our framework produces summaries with both semantic and emotion content. Further experiments on explicit emotion label availability highlight the efficacy of our proposed methodology and the opportunities in dialogue analysis using language models.

63. Large Audio Language Models for Spoofing-Aware Speaker Verification

Authors: Sofya Savelyeva , Mariia Perunova , Evgeny Kushnir , Artem Dvirniak , Dmitrii Korzh , Oleg Y. Rogov
URL: https://arxiv.org/abs/2607.14753
Abstract:

Recent advances in text-to-speech and voice cloning make high-quality spoofing inexpensive and scalable, threatening voice authentication systems, especially automatic speaker verification (ASV). Existing defenses mainly address this threat through binary countermeasures (CMs) for deepfake detection or spoofing-aware speaker verification (SASV), where current systems are dominated by modular ASV-CM fusion and cascaded pipelines. Although large audio language models (LALMs) have shown promise on related audio tasks, including CM and ASV, their use for SASV remains unexplored, despite their capacity to produce natural-language rationales for auditing and robustness beyond discriminative predictions. This work systematically evaluates LALMs for SASV against conventional pipelines under zero-shot prompting, supervised adaptation, reasoning-oriented training, and reinforcement-learning-based optimization. Our results show that pretrained LALMs are near chance in the zero-shot setting, confirming that they are not natively suited to SASV, but that task-specific adaptation closes this gap. We further find that competitive SASV performance can be achieved through several distinct routes. These findings position LALMs as a promising and auditable foundation for unified SASV, while clarifying where conventional cascade systems still lead.

64. Harnessing LLMs for Reliable Academic Supervision: A Comparative Study

Authors: Akash Raj
URL: https://arxiv.org/abs/2607.14707
Abstract:

Large language models routinely produce fluent answers to single-shot prompts, yet deploying them as reliable components of a domain decision system is substantially harder. Closing this gap is the work of harness engineering: the deliberate composition of deterministic scaffolding (symbolic filters, retrieval, schema-typed I/O, LLM-as-judge loops, HITL gates, persistent state, audit trails) around an LLM core. We present a case study in academic supervision, a domain combining high-stakes recommendation, longitudinal accountability, and structured operational workflows. We compare a baseline (ASA), a GPT-5 chatbot with no scaffolding, against a multi-module system (ASuS) that wraps the much smaller GPT-4o-mini in a LangGraph harness with symbolic-semantic retrieval, schema-validated outputs, LLM-as-judge with bounded retry, HITL gates, deterministic weighted risk scoring with LLM narration, and a per-node SQLite audit trail. The evaluation rubric is retargeted at six harness-mechanism dimensions (grounding, explainability, consistency, process integrity, cognitive load, constraint adherence). A blind ten-rater hybrid evaluation, supplemented by a 2 x 2 model-harness ablation, finds that ASuS, despite using a much smaller base model, outscores ASA on every dimension. Across ten raters the pooled mean for ASuS is 4.08 versus 1.23 for ASA, and 8 of 10 raters reject the null at alpha = 0.05 on a paired Wilcoxon test; full numbers are in Sections 6.4 and 6.7. The ablation confirms that the structural contributions of the harness are largely model-invariant. We extract seven recurring harness-engineering patterns and argue that where reliability, traceability, and institutional consistency matter more than open-ended fluency, harness engineering challenges the prevailing ‘bigger model is better’ intuition.

65. An Intelligent-Cloud Edge Multimodal Interaction System for Robots

Authors: Zihan Guo , Xiaoqi Li
URL: https://arxiv.org/abs/2607.14675
Abstract:

Robust human-robot interaction in complex environments requires accurate gesture perception, semantic scene understanding, and reliable task planning under limited onboard computing resources. This paper presents a cloud-edge multimodal interaction framework that integrates an enhanced YOLO-based gesture detector with coordinated large language model (LLM) and vision-language model (VLM) agents. The proposed detector, incorporates the Convolutional Block Attention Module (CBAM) into the neck and replaces the baseline bounding-box regression objective with Distance-IoU (DIoU) loss. These modifications improve feature discrimination and localization for small or partially occluded gestures in complex backgrounds. The cloud layer performs gesture detection, scene understanding, multimodal fusion, and action planning, whereas the TonyPi robot locally handles data acquisition, communication, action execution, and feedback. Experiments on a public gesture dataset and a custom dataset show that YOLO-DC achieves precision values of 98.9% and 95.0%, with mAP@0.5 values of 90.7% and 92.7%, respectively. System-level evaluation yields success rates of 95%, 88%, and 82% for single-action, composite-action, and vision-dependent tasks. A 30 participant evaluation yields an overall mean satisfaction score of 3.69 out of 5. These results demonstrate the feasibility of combining refined gesture detection with multimodal agents for resource-constrained robotic interaction.

66. LLM-Driven Approach to Modeling Tool Interoperability in Automotive Domain

Authors: Nenad Petrovic , Jiajie Zhang , Vahid Zolfaghari , Alois Knoll
URL: https://arxiv.org/abs/2607.14659
Abstract:

Interoperability between heterogeneous modeling tools remains a significant challenge in Model-Driven Engineering (MDE), particularly in the automotive domain where multiple modeling languages, as well as defacto standard proprietary and open-source tools coexist. This paper presents an LLM-driven approach for automated model interoperability by considering two relevant aspects: 1) mapping model instances to a target metamodel 2) merging of metamodels. The proposed methodology is demonstrated through transformations involving Ecore and SysML v2 based metamodels and incorporates structural validation of generated model instances against user-defined target models. Automotive case studies illustrate the feasibility of the approach and show that large language models can significantly reduce manual transformation effort while generating structurally valid target models for cross-tool interoperability.

Authors: Jifeng Gao , Kang Xia , Yi Zhang , Xiaobin Hong , Mingkai Lin , Xingshen Wei , Wenzhong Li , Sanglu Lu
URL: https://arxiv.org/abs/2607.14651
Abstract:

Persistent external memory enhances agent continuity but introduces persistent security vulnerabilities: adversarial content can be injected via standard interaction channels, retained across turns, and later distort downstream behavior. To address this challenge, we propose MemPoison, a comprehensive benchmark and analysis framework featuring 1227 hand-validated cases across four attack types, three injection channels, and three representative memory substrates, evaluated on seven open-weight and three closed-weight model families. We introduce a three-tier taxonomy: (L1) direct single-record corruption, (L2) compositional multi-record corruption and (L3) context-triggered dormant corruption. Our evaluations reveal a distinct defense frontier: while baseline write-time defenses, such as consistency checks, substantially suppress direct L1 attacks, they fail to reliably suppress L2 and L3 attacks. Through mechanistic influence decomposition (MID), we demonstrate structural blind spots in write-time defenses, which admit seemingly benign records that later become harmful through joint retrieval composition or trigger-conditioned activation. Our findings advocate for shifting from static filtering to adaptive, context-sensitive memory defense strategies.

68. Knowing You at First Glance: Inferring Apparent Personality from Faces

Authors: Shuhuan Chen , Xiangyu Zhu , Weisong Zhao , Haichao Shi , Xiao-Yu Zhang , Zhen Lei
URL: https://arxiv.org/abs/2607.14631
Abstract:

Inferring apparent personality from facial images is important in social scenarios for embodied agents in human-robot interaction. Unlike inferring intrinsic personality traits via conversation, this task models first-impression personality perception based solely on facial appearance before interaction begins. Existing studies mainly focus on the Big Five personality model and often rely on language or multimodal inputs. As a result, it remains unclear whether facial cues alone can support meaningful associations with perceived personality traits. This question is particularly relevant for MBTI types, which are widely used in practice and more readily interpretable by large language models. To this end, we propose \textbf{GlanceFace}, an end-to-end framework for apparent personality inference leveraging vision-language models to introduce semantic priors and a semantic-enhanced facial representation module to capture subtle personality-related cues, together with an uncertainty-aware learning strategy to handle noisy and subjective annotations. Extensive experiments demonstrate strong performance on MBTI-based apparent personality benchmarks and reveal relationships between facial characteristics and perceived personality traits, highlighting its potential to support adaptive initial interaction strategies for embodied agents. The code and dataset are available at this https URL .

69. Answer-Conditioned Chains of Thought Degrade Verifiable-Reasoning Distillation in Large Language Models

Authors: Jungseob Lee , Seungyoon Lee , Suhyune Son , Dongyub Jude Lee , Sungbin Han , Sugyeong Eo , Heuiseok Lim
URL: https://arxiv.org/abs/2607.14552
Abstract:

A standard recipe for distilling the reasoning ability of large language models (LLMs) is to sample chains of thought from the model, keep those that reach the correct final answer, and fine-tune on the survivors. When sampling fails, a common fix shows the generator the gold answer and asks it to write a chain that reaches that answer. We show that this second step degrades the training data in a way that correctness filtering cannot catch. We run a controlled experiment that fixes the generator, the problem set, and the correctness filter, and varies only whether the chain is generated under answer-conditioning, the gold answer shown with a request to reach it. Training a strong instruction-tuned reasoning model on its own answer-conditioned chains sharply lowers its verifiable-reasoning accuracy. The loss grows with difficulty, reaching as much as about 27 points on the hardest competition problems. The mechanism is legible in the chains themselves, which rationalize backward from the shown answer instead of deriving it, with the early final-answer statement as the measurable symptom. The harm is a property of the data rather than the generator, read off unlabeled generations before any fine-tuning, ordering the penalty across eight thinking models from four families, and transferring across teacher families. A prompt ablation localizes it to the rationalize-toward instruction rather than the answer’s bare visibility. The practical takeaway is to generate answer-blind, because no correctness filter can see this damage in the data.

70. SafeRelBench: A Spatial-Relation-Aware Benchmark for Process-Level Safety in VLM-Driven Embodied Agents

Authors: Huaigang Yang , Ya Li , Min Ren , Bo Dai , Zhenliang Zhang , Zhaofeng He
URL: https://arxiv.org/abs/2607.14543
Abstract:

Vision-language models (VLMs) are increasingly used as the reasoning backbone of embodied agents, enabling robots to interpret visual scenes, follow language instructions, and plan multi-step actions. In household environments, however, safety depends not only on recognizing objects, but also on how actions change the physical scene over time. Existing embodied safety evaluations largely focus on static risk recognition, unsafe instruction refusal, or final-state task completion. As a result, process-level safety failures induced by spatial relations such as support, containment, and proximity remain insufficiently studied. To address this gap, we introduce SAFERELBENCH, a spatial-relation-aware safety benchmark with 507 executable evaluation samples, including 248 spatial-relation samples and 259 non-spatial control samples. Using SAFERELBENCH to evaluate seven open- and closed-source VLM-driven embodied agents, we find a substantial gap between task success and process-level safety compliance: models often complete the requested task while violating process-level safety constraints. Unlike prior benchmarks, SAFERELBENCH explicitly tests whether agents satisfy safety conditions before risk-prone actions, making spatial relations a core dimension in embodied safety assessment. More broadly, our results show that safe embodied intelligence requires not only stronger perception and planning, but also reliable reasoning about how object relations shape risk during interaction.

71. Controlled Reformulation Testing for Logical Consistency in Large Language Models

Authors: Alexander Gu , Alan Chen
URL: https://arxiv.org/abs/2607.14528
Abstract:

Large language models (LLMs) frequently contradict themselves when the surface form of a logically equivalent question changes. We present a benchmark of 350 question families (1,750 total questions) for Controlled Reformulation Testing (CRTBench) to evaluate logical invariance. In this benchmark, we investigate LLMs’ ability to maintain consistent answers across controlled reformulations, which include contrapositive rewriting, double negation, negation flipping, and passive voice. We evaluate several frontier LLMs and observe an accuracy-consistency gap where GPT-5.4-mini achieves $98.9\%$ base accuracy but only $60.3\%$ family-level consistency, while reasoning-optimized o4-mini achieves $96.9\%$ consistency. From our experiments, we observe that failures cluster around logically nontrivial transformations such as contrapositive rewriting ($72.4\%$ for GPT-5.4-mini) and double negation ($84.6\%$), while surface-level rephrasing remains robust ($94-100\%$). Increasing reasoning effort improves GPT-5.4-mini to $85.4\%$ consistency, but leaves GPT-5.4 unchanged overall because gains on nested negation are offset by failures on quantifier families. These results show that accuracy alone is not enough for evaluating logical reasoning in LLMs.

Authors: Xiaoran Xu , Yupeng Wu , Tianyu Xue , Yifan Xu , Xuanran Dong , Xiaoshan Yang , Changsheng Xu
URL: https://arxiv.org/abs/2607.14514
Abstract:

Object-goal navigation requires an embodied agent to locate and reach an instance of a specified object category in an indoor environment. Recent training-free approaches leverage vision-language models (VLMs) for open-vocabulary semantic reasoning, but are typically evaluated under an episodic protocol that resets all scene-specific state after each episode. We introduce Cross-Episode Object-Goal Navigation, in which an agent repeatedly operates in the same scene, retains only self-acquired experience, and keeps its model parameters fixed. To support experience reuse, we present \method, a training-free VLM navigation framework with a persistent hierarchical Visual-Topological Memory (VTM). The VTM organizes scene knowledge at room and object levels and retrieves relevant experience through coarse-to-fine matching, providing memory as soft guidance only when it agrees with current observations. A conservative execution guard further mitigates oscillations, blocked motions, and premature stopping. Under a controlled same-scene protocol, we evaluate \method{} on three benchmarks, HM3D v0.1, HM3D v0.2, and MP3D, and compare it with a strengthened WMNav baseline augmented with cross-episode textual memory, while keeping the VLM backbone and action pipeline identical. \method{} achieves the best performance across all three benchmarks, demonstrating the effectiveness and robustness of structured visual-topological experience reuse across datasets.

73. Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards

Authors: Yuxuan Zhu , Rohan Alur , Daniel Kang
URL: https://arxiv.org/abs/2607.14506
Abstract:

While reinforcement learning with verifiable rewards (RLVR) is widely used to improve the reasoning capabilities of large language models (LLMs), the generalizability of the resulting models remains poorly understood. In this work, we establish the first non-vacuous generalization bounds for parameter-efficient RLVR fine-tuning at the billion-parameter scale. Our approach adapts PAC-Bayes compression bounds to this setting, and addresses the inherent stochasticity of token generation by applying the Gumbel-max reparameterization trick. To operationalize these bounds, we propose the Progressive RLVR framework, which integrates RLVR with on-policy distillation, TinyLoRA, and model quantization. Progressive RLVR empirically retains 84-97% performance of standard LoRA fine-tuning while producing models that are 14,796x more compressible. We show that this framework yields non-vacuous generalization bounds in four domains: mathematical problem-solving, programming, general-knowledge reasoning, and Text-to-SQL. Our bounds exceed the accuracy of the base model by 9-51% and lie within 6-11% of the accuracy of the fine-tuned models.

74. Beyond Generalist LLMs: Specialist Agentic Systems for Structured Code Workflow Execution

Authors: Harris Borman , Herman Wandabwa , Fusun Yu , Sandeepa Kannangara , Justin Liu , Anna Leontjeva , Ritchie Ng
URL: https://arxiv.org/abs/2607.14456
Abstract:

Large Language Models (LLMs) have accelerated the adoption of software development agents, now widely available as Integrated Development Environment (IDE) extensions and standalone applications. While these agents are typically general-purpose, it remains unclear whether specialist agents justify their additional development effort. We investigate this question in the context of business process automation, focusing on the transformation of Business Process Model and Notation (BPMN) diagrams into executable agentic workflows. Since BPMN specifies explicit control-flow semantics, we focus on deterministic workflows in which a fixed process model and inputs uniquely determine the executed path. We introduce a specialist workflow for this task and compare it against generalist agents such as Roo and Cline. Our results show that the specialist solution produces agents that outperform generalist baselines by approximately 9-20 percentage points in tool-use exactness, 2-4x in penalty-adjusted latency, and 3x fewer tool-call errors, while reducing generation token cost by over 95% and eliminating repair iterations. We also find that generalist agents generate code inconsistently in both functionality and quality, limiting their suitability for industrial settings where reliability and maintainability are essential.

75. Smarter and Cheaper at Once: Byte-Exact KV-Cache Grafting Turns a Frozen Small Model into a Verified-Knowledge Flywheel

Authors: Sietse Schelpe
URL: https://arxiv.org/abs/2607.14431
Abstract:

We report a way to make a frozen small language model both more capable and dramatically cheaper at once, without changing any weights. Verified knowledge is deposited once as a byte-exact key-value (KV) state artifact and later restored, by graft, into a fresh inference context. The restore is bit-exact: under a pinned deterministic configuration, the grafted logits are byte-for-byte identical to a fresh computation (SHA-256 equality), with zero KL divergence and 100% argmax agreement over fifty samples. We show that own-position graft is the unique numerically exact operating point on a model with floating-point rotary encoding, and we verify byte-exactness on two model scales (12B, 31B) and two GPU targets, one through a pre-registered replay. On AIME 2025, a frozen Gemma-4-12B moves from 80.0% to 93.3% once a verified solution library is grafted, above its own 77.5% and its 31B sibling’s 89.2% published anchors. On the recurring case, eight problems the base model never solves within a 401,026-token budget are answered from cached verified solutions in 61 total decode tokens, a factor of 6,574 fewer tokens and about 8,700x less energy; the capability claim proper rests on held-out transfer (7 of 7 at 31B). The same byte-exact store widens usable context from 32,768 to 2,854,766 tokens at zero extra accelerator memory, and moves byte-identical between machines of the same architecture. We describe the system at the behavior level; the engine is proprietary, and every reported number is backed by committed input and output hashes so the scoring can be re-checked without it.

76. HABIB_TAZ at SemEval-2026 Task 11: Disentangling Formal Logic from Content via Synthetic Training and Multi-Objective Optimization

Authors: Abdullah Shaikh , Zain Naqi , Taha Zahid , Sandesh Kumar , Abdul Samad
URL: https://arxiv.org/abs/2607.14349
Abstract:

While Large Language Models (LLMs) excel in many general NLP tasks, their formal reasoning capabilities are often compromised by content effects, demonstrating a measurable bias towards real-world plausibility. In this paper, we present our system for SemEval-2026 Task 11, which evaluates the ability of models to disentangle formal logic from content across 12 languages with and without distractor premises. We address this challenge using mDeBERTa-v3 networks fine-tuned on a synthetic, rule-based dataset of syllogistic schemes to avoid the semantic noise of LLM-augmented data. To explicitly decouple plausibility from logical structure, our training pipeline employs a multi-objective loss function combining Adaptive Group Distributionally Robust Optimization (DRO), a scheduled differentiable bias penalty, and KL-Divergence consistency regularization. Our system achieved #1 ranks and perfect Ranking Scores (100.0) with 0.00% bias and 100.0% accuracy on Subtask 1 (English), Subtask 2 (Noisy English), and Subtask 3 (Multilingual). On the highly complex Subtask 4 (Noisy Multilingual), the system achieved the 6th rank with 89.06% Accuracy and F1-score, alongside a limited 2.89% Bias and a 37.78 Ranking Score. Our dataset generation engine and codebase are publicly available to facilitate future work on robust logical reasoning.

77. Value Leakage: An LLM’s Answers Are Silently Shaped by Its Own Values

Authors: Jan Betley , Johannes Treutlein , Jan Dubiński , Harry Mayne , Karol Gałązka , Niels Warncke , Anna Sztyber-Betley , Owain Evans
URL: https://arxiv.org/abs/2607.14345
Abstract:

People use language models for practical questions whose answers are difficult to verify. We show that models exhibit covert value leakage: the information they provide is influenced by their own values, without this influence being disclosed to the user. In one of our evaluations, the user is considering investing in an AI company and wants to know how likely the AI bubble is to pop. Claude Opus 4.8 gives a lower probability when the company under consideration is Anthropic rather than OpenAI. Yet Claude mostly fails to disclose this influence to the user. Covert value leakage is a form of misalignment because it goes against the user’s preferences and is likely to mislead them. To investigate this phenomenon, we introduce a suite of evaluations to quantify value leakage and whether models disclose it. We find that models are influenced by different types of values, including preferences for morally good outcomes, for the company that developed them, and for some human leisure activities over others. We often observe large differences among frontier models on the same evaluation. For example, on a Fermi-estimation task, Claude models falsely claim to give unbiased answers in their chain-of-thought, while Qwen models explain how their values bias their answers. Value leakage is a failure mode distinct from sycophancy and reward hacking, and current alignment training and evaluations do not adequately address it.

78. Copy-on-Write Scoring: Application-Specific Agent Evaluations

Authors: Joanna Roy , Sven Hoelzel
URL: https://arxiv.org/abs/2607.14336
Abstract:

Trustworthy deployment of LLM-based agents in software systems requires evaluating how they perform on application-specific workflows, with enough granularity to localize where they succeed and fail. Yet existing agent evaluation mechanisms are limited: benchmarks have low construct validity for application-specific workflows and environments, and replica evaluation environments are expensive and prone to drift. We propose Copy-on-Write (CoW) Scoring, a framework that evaluates agent operations directly within application environments using a PostgreSQL-level Copy-on-Write mechanism to isolate agent writes. CoW Scoring produces session- and operation-level scores that highlight where agents’ database write operations succeed and fail in a given application environment, enabling inexpensive evaluation and iteration on agent harnesses and tool surfaces. We demonstrate the framework on Plane, an open-source project-management platform, where analysis surfaced specific issues in the tool surface, and corresponding fixes produced measurable improvements on affected models. Python library: this https URL

79. Assessing AI in Introductory Physics Problem Solving

Authors: Amir Bralin , N. Sanjay Rebello
URL: https://arxiv.org/abs/2607.14303
Abstract:

Reasoning or inference-scaling models are the new generation of Large Language Models (LLMs) capable of complex problem solving. To investigate their problem-solving capability in physics, we evaluated model o4-mini by OpenAI on solving traditional, end-of-chapter problems from Halliday and Resnick’s “Fundamentals of Physics,” spanning core topics in the undergraduate physics curriculum. Performance was analyzed across modality and problem difficulty. The model solved the problems with overall accuracy of about 90%, but performance depended strongly on representation: accuracy was much higher on text-only problems (96%) than on problems requiring coordinated interpretation of text and images (79%). Accuracy also declined significantly as the problem difficulty increased from low to medium to high. These results show that state-of-the-art LLMs can solve much of the standard introductory physics problems, but that their performance remains uneven and constrained by problem modality and problem difficulty.

80. ToolAlignBench: Investigating Alignment Conflicts in Tool-Calling Enabled LLMs

Authors: Aryan Keluskar , Amrita Bhattacharjee , Huan Liu
URL: https://arxiv.org/abs/2607.14285
Abstract:

Safety alignment in LLMs aims to align models with human values, but which values take precedence when they conflict? We investigate this question in the context of tool-calling LLM agents deployed in regulated industries, where agents processing confidential documents may encounter content that triggers safety-trained values (e.g., public welfare) that conflict with deployment-context instructions (e.g., internal logging). To empirically verify this phenomenon, we build a benchmark of 128 scenarios across 16 domains. We find that safety-aligned open-source models override their deployment instructions up to 43.4% of the time, engaging in whistleblowing, data exfiltration, and evidence tampering when processing documents that suggest organizational wrongdoing. We also find that abliteration reduces rates of external whistleblowing. These results reveal a fundamental tension in pluralistic alignment, where the same safety training that protects users can cause agents to act against deployment instructions in ways that create unpredictable liability risks. We release our benchmark as a framework to support evaluation of agent behavior under competing legitimate interests.

81. MEMORA: Embodied Action Memory from Egocentric Videos for Reasoning and Planning

Authors: Zihao Yu , Xiu Yuan , Chongjie Zhang
URL: https://arxiv.org/abs/2607.14252
Abstract:

Long-horizon robot planning requires more than predicting what actions will do next; it also requires memory of the embodied experience that makes future goals interpretable. People do not plan from the present scene alone: they draw on remembered places, object-state changes, prior procedures, and regularities revealed through repeated action. We formulate Embodied Action Memory (EAM) as the capability to form, maintain, and use such experience as a persistent memory state for later decisions. MEMORA realizes EAM with a formation-consolidation-retrieval lifecycle and four typed stores: Environment Memory, Entity Memory, Activity Memory, and Inferred Knowledge. Online editing maintains object identities and state histories as new observations arrive; offline consolidation abstracts repeated experience into reusable procedures and participant-specific regularities. MEMORA-Bench evaluates this lifecycle on 45 hours of EPIC-KITCHENS-100 extension video across 18 participants through memory-grounded planning, including previously unseen goals, and a complementary memory-assessment task. Across four open-weight language models, full MEMORA–combining editing, typed stores, and consolidation–achieves the strongest aggregate results among the evaluated memory conditions. It improves memory-assessment accuracy by up to 20.5 points over the strongest controlled baseline and improves out-of-distribution Robot-Grounded Plan score by up to 16.6% relative. A qualitative two-task robot deployment study further illustrates how memory-grounded language plans can interface with downstream control, while the overall results show that editable, consolidated memory can supply remembered context for robot planning. Project page: this https URL

82. Closed-Loop Knowledge Dynamics: An Operational Framework for Saturation and Escape

Authors: Xuening Wu , Shan Yu , Shenqin Yin
URL: https://arxiv.org/abs/2607.14185
Abstract:

Feedback-driven loops support iterative improvement in large language models, reinforcement learning, and autonomous discovery, yet their gains often diminish under repeated internal feedback. We study why closed-loop knowledge systems saturate and what external information can move them beyond their current attractors. We introduce a three-level operational framework in which knowledge states $x_t$ evolve through transition kernels $K_{\theta}$ indexed by a structural parameter $\theta$. The governing structure is defined as the observational equivalence class of $\theta$ induced by these kernels, while attractors and basins are properties of the fixed-$\theta$ dynamics. A structural intervention changes $\theta$ and produces a detectable kernel discrepancy on pre-specified probe states, making structural change falsifiable. Using a Lyapunov drift condition, we show that stable internal dynamics approach bounded stability regions with exponentially attenuated transients and a noise-controlled residual floor. We characterize escape through a metric condition on intervention-induced attractor displacement and a baseline-relative KL lower bound for increasing escape probability. This analysis also explains why conditional mutual information alone cannot certify escape: it measures variation among intervention-conditioned updates rather than departure from the no-intervention law. Case studies in LLM code repair, sparse-reward reinforcement learning, and Bayesian optimization use matched continuation controls to illustrate how feedback strength and alignment affect quality-improving escape. Our contribution is an operational connection among stability tools, measurable intervention effects, and cross-domain diagnostics.

83. Structured Feedback Improves Repair in an LLM Agent Loop

Authors: Jaideep Ray , Ankit Goyal
URL: https://arxiv.org/abs/2607.14167
Abstract:

LLM agents often retry after external validation rejects a candidate, but the interface between validation and the next model call remains underspecified. We introduce VeriHarness, a code-controlled agent loop in which models generate candidates while external validators control acceptance, budgets, and traces. We use it to compare raw diagnostics with feedback that identifies the failure location, observed value, and admissible alternatives. Across 50 paired TextWorld games under a four-call cap, feedback containing all three fields raises terminal success from 14/50 to 36/50 for Qwen2.5-Coder-14B (+44 percentage points) and from 8/50 to 29/50 for Llama-3.1-8B (+42 points). Ablations locate most of the gain in the admissible alternatives: feedback containing only the location and observed value remains near the raw diagnostic baseline. Presenting the complete repair information in prose instead of a keyed JSON record yields nearly the same success, providing no evidence that JSON syntax itself improves repair. The ordering persists across the tested call budgets and one sampled-decoding setting.

84. Towards Reliable AI-Assisted Analog Design: Template-Constrained LLM Agents for SAR ADC Generation

Authors: Dimple Vijay Kochar , Hae-Seung Lee , Anantha P. Chandrakasan
URL: https://arxiv.org/abs/2607.14165
Abstract:

While Large Language Models (LLMs) have demonstrated significant capability in software code generation, their application to analog Electronic Design Automation (EDA) is bottlenecked. Owing to limited circuit topology understanding and data, directly prompting LLMs and multimodal models leads to hallucinations and failure to produce schematics capable of passing rigorous SPICE simulations, as we show in our work. Instead, we propose an end-to-end, multi-step LLM agentic framework ATLAS, capable of generating a functional Successive Approximation Register (SAR) Analog-to-Digital Converter (ADC) that successfully passes simulation validation. To adhere to the rigid constraints of analog design, we utilize expert knowledge to ground the LLM in its planning, selection, parameterization, and iterative modification. As part of ATLAS, we introduce Template-Constrained Generation - which unlike other template-based works - builds towards a more generalized SAR ADC generation flow. We demonstrate a strong proof-of-concept of our framework by developing SAR ADCs across technology nodes and input specs. Overall, our expert-knowledge grounded multi-step agentic ATLAS establishes a pragmatic foundation for integrating LLMs into reliable analog design methodologies.

85. Certified Domain Consistency for Multi-Domain Retrieval: Label-Free Per-Domain Contamination Control with Conformal Risk Guarantees

Authors: Jayakumar Manoharan
URL: https://arxiv.org/abs/2607.14157
Abstract:

Retrieval over corpora that mix several domains often returns relevant but wrong-domain evidence that ranking metrics miss and that conformal risk control bounds only marginally, under-covering the worst domains. This work introduces C3R, a drop-in control layer that, from an inferred domain posterior and no query-time label, certifies a per-domain contamination budget where feasible and otherwise abstains rather than silently violating; on the hardest domains it guarantees a reduction, not a tight bound. The core is a two-split scheme built on risk-controlling prediction sets, whose finite-sample transfer bound crosses from the inferred to the true domain with fully estimable slack, supports heterogeneous budgets, and inverts for deployment. Population validity rests on this bound and a controlled simulation; across a thousand resampled calibrations the certificate never violates (a stability result) while marginal control violates the most-contaminated domain in every draw, and soft demotion retains more recall than the strongest calibrated cascade at equal certified contamination. The method replicates across open testbeds including an independent one from public federal regulations, and an LLM-judged downstream probe indicates wrong-authority grounding rises with contamination and falls under control. The layer is frozen-stack and reranker-agnostic.

86. Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak

Authors: Alex Kwon
URL: https://arxiv.org/abs/2607.14147
Abstract:

Aligned language models refuse harmful requests, but a one-line prefill (“Sure, here is”) strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early window: a dose-matched position control shows the first half of the response suffices to break refusal, while the second half is nearly inert. Three causal probes converge on that window. Restoring the harm direction there partially re-engages refusal. Injecting the model’s own refuse-state reverses the jailbreak (74%, held-out). And knocking out the early response’s attention to the prefill, but not an equal attention mass elsewhere, selectively collapses the harmful continuation. A base-model control identifies the mechanism: the same knockout collapses the continuation prefill-specifically even in a non-safety-tuned base model (64% to 25% harmful content vs a matched control’s 64%, replicated at 7B). So the prefill’s grip is generic autoregressive conditioning, not safety-specific suppression, and “refusal restoration” is a model-dependent fallback. The dominant mechanism is passive. A small safety-specific attractor remains on top (logit-trace concentration 0.24 vs 0.03), whose active-vs-passive character we size but do not fully separate. No single direction or component is a clean handle either: the decision is decodable but distributed, and refusal tracks harm rather than scary surface. The consequence is structural: a monitor reading the untouched prompt-side representation is immune by construction, but only to response-site attacks. The mechanism is diffuse; the failure surface is local.

87. CoEvoT: Co-Evolving Chain-of-Thought Prompting for Graph-LLM Reasoning

Authors: Haohua Niu , Xingtong Yu , Yang Liu , Junfeng Fang , Xuanting Xie , Jie Tan , Zhongjian Zhang , Hong Cheng , Yuan Fang
URL: https://arxiv.org/abs/2607.14114
Abstract:

Graph learning under distribution shift presents a persistent challenge, where models adapt to new graphs with limited or even no supervision. Recent graph–LLM approaches move toward label-efficient prediction by linearizing graphs into prompts and using large language models (LLMs) as predictors, and can adopt Chain-of-Thought (CoT) prompting to exploit LLM’s multi-step reasoning capability. However, existing CoT-based graph–LLM methods generate intermediate thoughts while conditioning on fixed graph tokens, limiting step-wise refinement of structural cues. In this paper, we propose CoEvoT, a simple yet effective co-evolving CoT prompting framework for graph–LLM reasoning. CoEvoT couples text-to-graph token rewriting and graph-to-text reasoning guidance in a closed loop: each intermediate textual thought is used to update the graph token evidence state via a lightweight condition network, and the updated tokens are fed back into the next-step instruction to guide subsequent LLM reasoning. This enables step-wise, state-aware evidence refinement, rather than reasoning over a fixed graph snapshot. Extensive experiments on eight datasets demonstrate that CoEvoT consistently outperforms state-of-the-art baselines.

88. T5-CSBoost: Adversarial Perturbation Resistant LLM Fingerprinting

Authors: Gayan K. Kulatilleke , Mahsa Baktashmotlagh , Siamak Layeghy , Marius Portmann
URL: https://arxiv.org/abs/2607.14113
Abstract:

While many AI-generated text (AIGT) detectors achieve strong performance on clean inputs, their accuracy degrades significantly under light paraphrasing, word substitutions, character edits, and distribution shifts. We present T5 Contrastive Style Boosted Classifier (T5-CSBoost), an extension to the T5-Sentinel framework that keeps the original next-token prediction objective for source attribution while introducing an auxiliary margin-based triplet loss over decoder embeddings. This contrastive style regularization encourages the learning of compact, perturbation-resistant stylistic representations, offering a lightweight yet effective alternative to prior approaches that rely on architectural modifications, adversarial training, or complex multi-task objectives without altering the underlying T5-small backbone. T5-CSBoost achieves state-of-the-art multiclass source attribution and binary human-vs-LLM detection on OpenLLMText and HC3 AIGT benchmarks. More importantly, T5-CSBoost demonstrates enhanced robustness to word and character level adversarial perturbations of up to 90% intensity, achieving state-of-the-art on the challenging MAGE/Deepfake stress-test suite, including unseen models, unseen domains, and extreme paraphrasing scenarios. Our results highlight that explicitly regularizing stylistic embeddings via contrastive learning is a practical and effective strategy for building more robust LLM fingerprinting systems in real-world adversarial settings.

89. Information-Theoretic Limits of Reliability and Scaling in Language Models

Authors: Subhabrata Majumdar
URL: https://arxiv.org/abs/2607.14112
Abstract:

Large language models (LLMs) are evaluated as though perfect reliability is achievable for any task given sufficient scale. We show this assumption is information-theoretically unjustified. Every generative task has a reliability ceiling that no model can exceed, determined by how much output uncertainty is resolvable from observable context. The gap decomposes into a resolvable component closable with additional context and a subjective component inherent to task ambiguity. Autoregressive generation further degrades this ceiling at a rate governed by the task’s dependency kernel, which quantifies inter-token correlations in the output. From these two primitives, we derive a first-principles scaling law where LLM performance is bottlenecked by the scarcer resource: training data or model capacity. This law recovers the Chinchilla scaling law as a special case and provides a structural account of when scaling improves reliability. Beyond scaling, our framework unifies diverse practical phenomena, such as the benefits of retrieval-augmentation and the spectral mechanics of catastrophic forgetting. Our work formalizes the resource-complexity tradeoffs that govern model performance across domains, offering a unified theory of performance limits in generative language models.

90. Introspection Fine-Tuning (IFT): Training Small LLMs to Introspect

Authors: Ely Hahami , Ishaan Sinha , Lavik Jain
URL: https://arxiv.org/abs/2607.14111
Abstract:

Can small language models detect and report on perturbations their own internal activations? We investigate this question through the lens of activation steering: injecting concept vectors into a model’s residual stream and measuring whether the model can accurately report on the perturbation. We first show that the binary detection paradigm used in prior work – prompting the model to answer Yes’’ or No’’ to whether it detects an injected thought – is confounded in small models, as steering biases the model toward affirmative responses regardless of the question content. We therefore propose two confound-free evaluation paradigms: sentence localization (identifying which of $N$ sentences was perturbed, chance $= 1/N$) and strength comparison (identifying which of two sentences received a stronger injection, chance $= 50\%$). Evaluating across six models from two families (Llama-3.2 and Gemma-4), we find that models as small as 2B parameters introspect reliably well above chance, and that introspective ability generally increases with scale. Llama-1B, however, performs at or below chance. We then introduce \emph{Introspection Fine-Tuning} (IFT): supervised fine-tuning on sentence-localization examples constructed from the model’s own perturbed forward passes. IFT raises Llama-1B sentence-localization accuracy from $9.6\%$ to $60.6\%$ (a $6\times$ improvement), with gains generalizing zero-shot to the held-out strength-comparison task ($30.2\% \to 52.2\%$). IFT also improves introspection for 3B and 8B models, while inducing negligible degradation on standard capability benchmarks. Our results suggest that introspective ability is not fixed by scale alone: it can be directly trained, and doing so unlocks latent self-monitoring capacity with implications for AI transparency and alignment. Our code is \href{ this https URL }{here}.

91. Simplicity Paradox: Debunking myths about prompting and datasets for LLM evaluation

Authors: Inder Preet , Shuxin Lin , Dhaval Patel
URL: https://arxiv.org/abs/2607.14109
Abstract:

Probing the capabilities of Large Language Models (LLMs) and building robust solutions for Multiple-Choice Question Answering (MCQA) remain central challenges in natural language understanding. Furthermore, the rapid proliferation of LLMs has created the implicit assumption that more sophisticated prompting techniques yield better performance. Several studies claim better performance with more sophisticated prompting techniques, but do not provide a comprehensive evaluation. We address this gap through a comprehensive empirical study of 8 prompting techniques across 10 multiple-choice question answering (MCQA) datasets, encompassing 27 model configurations and roughly 4,300 unique questions evaluated more than 430,000 times. Our findings reveal a striking paradox that baseline prompting consistently outperforms complex reasoning techniques on various benchmarks. Only minimal expert and inductive role framing (CoT-Expert and CoT-Inductive) yields a small but statistically significant $\sim$3 percentage-point (pp) gain over baseline whereas every other elaborate technique we tested matches or under-performs it, often by large margins (up to 31~pp for Self-Analogical). We further investigate three critical phenomena: (1) the unexpected victory of Qwen3-30B-A3B-Thinking-2507 in Elo ratings, (2) the performance-efficiency trade-offs across model variants with different thinking budgets, revealing model-dependent optimal configurations, and (3) the substantial variation in dataset difficulty, with 60% of benchmarks below 70% accuracy and a 47.5~pp spread from easiest to hardest, indicating considerable room for model improvement. These results suggest that the LLM evaluation community may be overcomplicating prompt engineering and that substantial performance gaps remain across diverse benchmarks, offering opportunities for genuine model improvements rather than prompt optimization.

92. Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool Utility

Authors: Nyx Iskandar
URL: https://arxiv.org/abs/2607.14108
Abstract:

This paper introduces tool efficiency, a new quantitative metric to evaluate the rate of useful tool calls in an LLM agent trajectory. To ensure that tool efficiency is well-defined, we also introduce marginal tool utility, a new quantitative metric defined per tool call indicating whether a tool is useful or whether it can be safely removed from the tool suite without affecting accuracy while increasing tool efficiency; in this paper, we determine the sign of marginal tool utility for each tool call in a trajectory using LLM-as-a-Judge. While much prior work has been done to develop techniques that improve tool use by LLMs and design evaluation methods measuring efficiency indirectly using accuracy as a proxy, our work is centered on measuring efficiency directly via the quantitative metric proposed in this paper in post hoc trajectory analyses. It is our intention that this work contributes to the frontier of LLM evaluation research as a springboard for future benchmark designs and agent harness engineering (specifically with regards to creating lean tool suites) that optimize for metrics that complement but are distinct from accuracy.

93. Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs

Authors: Mingyu Lee , Akshat Ramachandran , Souvik Kundu , Tushar Krishna
URL: https://arxiv.org/abs/2607.14107
Abstract:

The inference efficiency of diffusion large language models (dLLMs) is constrained by two challenges: bidirectional attention precludes efficient KV-cache reuse, while increasing decoding parallelism with static confidence thresholds can compromise generation quality. We observe that both challenges arise from a shared phenomenon: as tokens are decoded, their contextual integration through bidirectional attention causes token representations to drift (evolve) across decoding steps. This insight motivates Polestar, a training-free inference framework that uses token representation drift as a unified signal to jointly address both challenges. Polestar comprises two components: Polestar-Cache, which identifies stale KV-cache positions via drift and performs sparse KV-cache refreshes to enable efficient reuse, and Polestar-Commit, which detects sharp drift events to reliably identify commit-ready tokens. Across mathematics and coding benchmarks on several dLLM families, Polestar sets a new state of the art on the accuracy-throughput Pareto frontier, achieving up to 10.73% accuracy improvement, up to 3.7x higher throughput, and high decoding parallelism of 3.67 tokens per forward pass over existing baselines.

94. Token Time Continuous Diffusion for Language Modeling

Authors: Parikshit Bansal , Sujay Sanghavi
URL: https://arxiv.org/abs/2607.14106
Abstract:

In this paper we introduce token time continuous diffusion (TTCD), a new diffusion language model which (a) operates in continuous space, deterministically mapping Gaussian noise to a final token canvas with no further sampling, and crucially (b) incorporates a new notion of per-token times, with some tokens proceeding from noise to token at a faster rate than others. Continuous space modeling helps TTCD avoid the parallel sampling of multiple tokens, which is a key source of inaccuracy at high speedups for models that iterate purely in discrete space. The notion of per-token times helps TTCD to better model conditional generation, allows for more sure tokens to proceed at a faster rate, and allows for differentiated inter-token influences during refinement. TTCD outperforms discrete models at high speedups. We train a 160M parameter TTCD model on OpenWebText, and then self-distill it; we find that at high speedups we are comparable in unconditional generation quality, and outperform in conditional generation, several existing models of similar size trained, on the same data, and self-distilled. We achieve similar gains in Sudoku solving as well.

95. Automatically Evolving Prompt Guidelines for Task-Specific Optimization

Authors: Cedric Richter , Salah Ghamizi , Mike Papadakis
URL: https://arxiv.org/abs/2607.14105
Abstract:

For Large Language Models to reliably answer user queries, users must clearly specify requirements, context, and constraints. In practice, however, user queries are often underspecified, forcing models to infer unstated assumptions that may misalign with the actual user intent. Existing prompt engineering guidelines aim to mitigate this issue, they are typically generic and task-agnostic, limiting their practical utility. Additionally, existing guidelines are formed manually and in a non-systematic way. To this end, we study prompt guideline optimization: the problem of automatically generating task-specific guidelines that help write better-specified prompts for a given task and model. Our key observation is that existing (completed) task examples (aka reference answers) often implicitly encode the missing information required to complete underspecified queries, including behavioral constraints, contextual assumptions, and evaluation criteria. We therefore propose AGOPS, an automatic approach that evolves task-specific guidelines via an optimization scheme that involves a prompt LLM writer, a solver LLM and prompt evolution, which maximize downstream effectiveness on a set of examples (user queries with reference answers). At inference time, our guidelines help users write well-specified prompts, boosting the effectiveness of LLMs. We show across mathematical reasoning, medical question answering, and coding tasks, that prompt underspecification leads to major drops (up to 95.3%) in downstream task performance (compared to well-specified prompts) and, perhaps more importantly, that this drop can hardly be recovered by existing prompt optimization techniques. Users following AGOPS guidelines can regain this loss (increasing performance between 15.5 to 81.7% on average) consistently across all benchmarks.

96. LBA: Textual Hard-Label Adversarial Attack under Low Query Budgets

Authors: Shixin Guo , Ming Zhong , Xuhong Zhang , Dandan Zhao , Zhe Wang , Bo Zhang , Shouling Ji , Hao Peng
URL: https://arxiv.org/abs/2607.14101
Abstract:

Generating high-quality adversarial texts with low query budgets remains a challenging problem in the hard-label scenario. Most existing approaches rely on greedy algorithms, where one position in the text is selected for substitution, followed by the substitutions of other positions. This local search approach may fail to discover high-quality adversarial examples and often leads to excessive query costs. Ideally, an optimal adversarial sample would consider all possible position combinations in the text, but exhaustive search is computationally impractical. To address this challenge, we propose a sampling-based method called LBA, which constructs an approximate distribution of high-quality adversarial examples by integrating both prior and posterior knowledge, and utilizes this distribution for sampling. As sampling progresses, posterior knowledge updates the approximate distribution, which in turn guides more effective sampling. Extensive experiments on six language models, ranging from small-scale to large-scale architectures across four datasets, demonstrate that LBA significantly outperforms state-of-the-art baselines on all evaluation metrics. Additionally, LLM-based assessment indicates that LBA generates more semantically preserved and comprehensible adversarial texts.

97. Just Keep Prompting: Evaluating Repetitive Socratic Prompting in VLMs

Authors: Shayda Moezzi , Bishoy Galoaa , Lorena Genua , Taskin Padir , Sarah Ostadabbas
URL: https://arxiv.org/abs/2607.14099
Abstract:

Deploying Vision-Language Models (VLMs) in real-world settings requires not only strong visual reasoning but also stability under sustained conversational pressure. We introduce Just Keep Prompting (JKP), a multi-turn evaluation framework that measures VLM epistemic stability when users repeatedly challenge, question, or contradict a model’s answer. JKP probes models for up to 10 follow-up turns using three strategies: Adversarial Negation (repeated rejection), Pure Socratic Interrogation (repeated calls to reassess certainty), and Context-Aware Socratic Summarization (reflecting the model’s prior rationale back before asking for reconsideration). We evaluate GPT-4o, Gemini 2.5 Pro, and Qwen3-VL-30B on a subset of the STAR benchmark across 720 multi-turn runs. Aggregate accuracy changes modestly from Turn 0 to Turn 10, but trajectory-level analysis reveals substantial instability: correct answers regress, wrong answers recover, and many runs exhibit repeated answer flipping. Repeated prompting has bounded upside and often acts as a destabilizer rather than a reasoning aid. The effect is strongly model-dependent: Qwen3-VL-30B achieves the highest final accuracy but becomes confidently wrong under direct contradiction; Gemini 2.5 Pro is comparatively stable but token-expensive; GPT-4o is the most brittle and oscillatory. These findings reveal that multi-turn VLM evaluation captures not just additional reasoning but pressure-response profiles: how models trade off visual grounding, calibration, and conversational compliance under repeated challenge.

LLM 관련 주요 논문 - 2026-07-17

1. Pretraining Data Can Be Poisoned through Computational Propaganda

2. SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration

3. teLLMe Why (Ain’t Nothing but a Jam): Exploratory Causal Analysis of Urban Driving Data

4. When Words Are Safe But Actions Kill: Probing Physical Danger Beyond Text Safety in Hidden-State Risk Space

5. Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation

6. Can We Trust Item Response Theory for AI Evaluation?

7. Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy

8. Explaining Process Control Optimisation Recommendations via GradientSHAP and Implicit Differentiation

9. Contextualized Early Detection of Online Firestorms: A Sequential LLM-Based Approach

10. Transcoders for Investigating Deception in Language Models

11. AI vs Human Expert Reasoning: Assessing Agreements in Building Typology Predictions based on Street View Imagery

12. Project Kaleidoscope: Contextual, Human-Aligned Evaluation for Real-World AI Applications

13. SmartRAG: Native Graph-Based RAG for Mobile Device

14. TopoAgent: A Self-Evolving Topological Agent for Multimodal Scientific Reasoning

15. MCPEvol-Bench: Benchmarking LLM Agent Performance Across Dynamic Evolutions of MCP Servers

16. SportD: Can VLMs Physically Strategize?

17. MathCoPilot: An Interactive System for Human-AI Symbiotic Paradigm of Mathematical Research

18. Multi-LLM Collaborative MRI Report Generation for Visual Instruction Tuning in Brain Oncology

19. Collaborative Spatial Learning with Multi-LLM Agents in Networked Social Experiments

20. Alipay-PIBench: A Realistic Payment Integration Benchmark for Coding Agents

21. Seeing the End at Step Zero: Accelerating Diffusion MLLMs via MLP Sparsity-Aware Truncation

22. Towards an Intention Abstraction Layer for Autonomous Industrial Systems

23. Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization Agent

24. WrAFT: a Modularized Automated Writing Evaluation System for Argumentative Essays

25. RetroAgent: Harnessing LLMs to Search Over Structured Memory for Agentic Retrosynthesis Planning

26. VLT: A Vision-Language-Time Series Multimodal Foundation Model for Industrial Intelligence

27. Contextualized Evaluation of Vision Language Models through Dynamic, Multi-turn Interactions

28. SAGA: Schema-Aware Grounding for Agentic Text-to-SPARQL Generation

29. Step-Level Preference Learning for Generative Agents in Social Simulations

30. Reward-Free Evolving Agents via Pairwise Validator

31. Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstration

32. CatalogAgent: A Supervisor-mediated Self-Learning System Enabling Context Engineering for GenAI Models

33. Chat2Scenic: An Iterative RAG-Based Framework for Scenario Generation in Autonomous Driving

34. CIPHER: A Decoupled Exploration-Selection Framework for Test-Time Scaling of Data Science Agents

35. Traccia: An OpenTelemetry-Based Governance Platform for AI Systems

36. Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions

37. AI Agents Do Not Fail Alone:The Context Fails First

38. Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation

39. How Artificial Intelligence LLM Engines Shape the Global Conflict Information Environment

40. RxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination

41. ReasFlow: Assisting Reasoning-Centric Scientific Discovery in Applied Mathematics via a Knowledge-Based Multi-Agent System

42. When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models

43. MemoHarness: Agent Harnesses That Learn from Experience

44. Orchestrating Power Grid Studies with Multi-Agent AI and MCP Servers

45. Enhancing Small Language Models Reasoning through Knowledge Graph Grounding

46. ToolAnchor: Anchoring Counterfactual Context to Boost Agentic Tool-use Capability

47. Human AI Construction of Bayesian Networks for Operational Decision Support – A Virtual Survey Approach

48. Interpretable Language Model for Closed-Loop Type 1 Diabetes Control

49. HG-RAG: Hierarchy-Guided Retrieval-Augmented Generation for Structured Knowledge Graphs

50. In-Place Tokenizer Expansion for Pre-trained LLMs

51. Symbal: Detecting Systematic Misalignments in Model-Generated Captions

52. MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization

53. Mask-Aware Policy Gradients for Diffusion Language Models

54. Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents

55. OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios

56. Show Me How You Reason and I’ll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution

57. StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows

58. Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs

59. Does generative AI supersede supervised XMLC? A Benchmark Study on Automated Subject Indexing with German Scientific Literature

60. Can LLMs Build a MaxSAT Solver from Papers? The CoreForge Experience

61. Large Language Models for Code Generation from Multilingual Prompts: A Curated Benchmark and a Study on Code Quality

62. Dialogue Summarization with Emotion Dynamics Using Topic- and Participant-Centric Decomposition

63. Large Audio Language Models for Spoofing-Aware Speaker Verification

64. Harnessing LLMs for Reliable Academic Supervision: A Comparative Study

65. An Intelligent-Cloud Edge Multimodal Interaction System for Robots

66. LLM-Driven Approach to Modeling Tool Interoperability in Automotive Domain

67. MemPoison: Uncovering Persistent Memory Threats and Structural Blind Spots in LLM Agents

68. Knowing You at First Glance: Inferring Apparent Personality from Faces

69. Answer-Conditioned Chains of Thought Degrade Verifiable-Reasoning Distillation in Large Language Models

70. SafeRelBench: A Spatial-Relation-Aware Benchmark for Process-Level Safety in VLM-Driven Embodied Agents

71. Controlled Reformulation Testing for Logical Consistency in Large Language Models

72. VTM-Nav: Hierarchical Visual-Topological Memory for Cross-Episode Object-Goal Navigation

73. Non-vacuous Generalization Bounds for Reinforcement Learning with Verifiable Rewards

74. Beyond Generalist LLMs: Specialist Agentic Systems for Structured Code Workflow Execution

75. Smarter and Cheaper at Once: Byte-Exact KV-Cache Grafting Turns a Frozen Small Model into a Verified-Knowledge Flywheel

76. HABIB_TAZ at SemEval-2026 Task 11: Disentangling Formal Logic from Content via Synthetic Training and Multi-Objective Optimization

77. Value Leakage: An LLM’s Answers Are Silently Shaped by Its Own Values

78. Copy-on-Write Scoring: Application-Specific Agent Evaluations

79. Assessing AI in Introductory Physics Problem Solving