TechniqueRLHF / Alignment8 recent entries15 Oct 2026AuthGateBench: Benchmarking LLMs as Authorization GatearXiv:2610.16426v1 Announce Type: new Abstract: Automated approval systems let coding agents execute high-risk actions with minimal human oversight by delegating the authorization decision to a classi→15 Oct 2026Audit tools of the trade: Automated Auditor Elicitation Can Be Sensitive to the Context the Auditor Is GivenarXiv:2610.16911v1 Announce Type: new Abstract: Does providing additional context to automated AI auditors improve their effectiveness? Using the Petri tool to audit LLM coding agents, we test the imp
TechniqueRAG8 recent entries15 Oct 2026EGCForest: Structured Agent Memory Organization for Enterprise Group ChatsarXiv:2610.19836v1 Announce Type: new Abstract: Enterprise group chats record project progress, decisions, task ownership, and other organizational information valuable to enterprise agents. Reliable →15 Oct 2026DTScribe: Generating Models for Digital Twin Reporting using Local Large Language ModelsarXiv:2610.18635v1 Announce Type: cross Abstract: The development of a digital twin brings together many stakeholders and technical components. We have previously investigated the reporting of the con→15 Oct 2026DarKnight: Retrieval-Guided Agentic Framework for Breaking Dark Web CAPTCHA GatesarXiv:2610.19830v1 Announce Type: cross Abstract: Cyber Threat Intelligence (CTI) investigators who monitor dark web marketplaces for leaked credentials and breach listings must pass the diverse CAPTC→15 Oct 2026Confidently Wrong: Detecting Hallucinations in Financial Question Answering from LLM Internal StatesarXiv:2607.11414v2 Announce Type: replace Abstract: Large language models (LLMs) in financial applications fail most consequentially when they are confidently wrong. Hedged, uncertain answers invite s→15 Oct 2026Can LLMs Access their World Knowledge for Event Prediction?arXiv:2610.19021v1 Announce Type: new Abstract: Recent work has questioned whether large language models (LLMs) can be useful for temporal event prediction. Despite vast pretrained world knowledge, LL→15 Oct 2026AutoSec-RAN: Specification-Grounded LLM Fine-Tuning for Automated Security Analysis of O-RANarXiv:2610.15995v1 Announce Type: cross Abstract: This paper presents AutoSec-RAN, a specification-grounded large language model (LLM) framework for automated security triage of Open Radio Access Netw→15 Oct 2026ASIWorld: Benchmarking Autonomous Agents in a Persistent Survival SandboxarXiv:2610.14807v1 Announce Type: new Abstract: Evaluating autonomous agents requires examining whether they can translate goals into coherent action across extended interactions. In persistent enviro→15 Oct 2026ActICL: Post-Training for Agentic Multimodal In-Context LearningarXiv:2610.19475v1 Announce Type: cross Abstract: Multimodal in-context learning (ICL) is an important capability of multimodal large language models (MLLMs). Its performance is closely tied to the de
TechniqueAgents8 recent entries15 Oct 2026ActICL: Post-Training for Agentic Multimodal In-Context LearningarXiv:2610.19475v1 Announce Type: cross Abstract: Multimodal in-context learning (ICL) is an important capability of multimodal large language models (MLLMs). Its performance is closely tied to the de→15 Oct 2026ACED-Bench: Evaluating How LLM Agents Acquire and Act on Causal EvidencearXiv:2610.14381v1 Announce Type: new Abstract: Large language models are increasingly used as scientific assistants, yet we still lack reliable ways to evaluate whether they can investigate a system →15 Oct 2026ABot-ER: Unified Embodied Reasoning through Agentic Data-Model Co-EvolutionarXiv:2610.18355v2 Announce Type: new Abstract: We introduce ABot Embodied Reasoning (ABot-ER), a unified embodied reasoning foundation model that breaks the tension between breadth and depth in embod→15 Oct 2026A Strong Harness Narrows the Model Gap: A Comparative Evaluation for Legislative Document-to-XML ConversionarXiv:2610.19999v1 Announce Type: new Abstract: Legal-content reproduction requires matching the source exactly in section hierarchy, verbatim text, citations, and amendment spans; a dropped citation →15 Oct 2026A Sovereign, Indic-First, Reasoning ModelarXiv:2610.15371v1 Announce Type: cross Abstract: Gnani-Evon-v3.3-30B-A3B is a long-context hybrid Mamba-Transformer mixture-of-experts model for reasoning, instruction following, and tool use in Engl→15 Oct 2026A Generalist Hanabi AgentarXiv:2503.14555v2 Announce Type: replace-cross Abstract: Traditional multi-agent reinforcement learning (MARL) systems can develop cooperative strategies through repeated interactions. However, these→15 Oct 2026A Decade in a Day: Synthesizing Agent-Native Interfaces from Human-Oriented Software EvolutionarXiv:2610.15765v1 Announce Type: cross Abstract: Large language model agents are emerging as an important new class of users of desktop applications. However, existing desktop interfaces were designe→15 Oct 2026A Counterfactual Graph Approach to Oppositional Narrative AnalysisarXiv:2603.06135v3 Announce Type: replace-cross Abstract: Distinguishing legitimate critical speech from coordinated conspiracy narratives is a pressing challenge for the integrity of the online infor
TechniqueFine-tuning8 recent entries15 Oct 2026Cross-Scenario Industrial Defect Detection via Domain-Robust Training and Metric-Aligned SelectionarXiv:2610.19540v1 Announce Type: new Abstract: Automated defect detection of industrial boards must remain highly reliable under cross-scenario variations in lighting, surface contaminants, and produ→15 Oct 2026Cowork-Gym: Synthesizing Interactive Work Environments at ScalearXiv:2610.17670v1 Announce Type: new Abstract: Cowork agents navigate dependencies across tools and files while adapting to changing user requirements and edits to shared workspaces. Scaling such env→15 Oct 2026Coupled Harness--Model Evolution for Scientific Equation DiscoveryarXiv:2610.17740v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to discover scientific equations from experimental data. Such a discovery system has two parts that m→15 Oct 2026Beyond Successful Play: Diagnosing Opponent Modeling in Large Language ModelsarXiv:2610.17310v1 Announce Type: new Abstract: Language-model agents are expected to adapt to others whose behavior changes through interaction. Yet adaptation may rely on learned behavioral patterns→15 Oct 2026AutoSec-RAN: Specification-Grounded LLM Fine-Tuning for Automated Security Analysis of O-RANarXiv:2610.15995v1 Announce Type: cross Abstract: This paper presents AutoSec-RAN, a specification-grounded large language model (LLM) framework for automated security triage of Open Radio Access Netw→15 Oct 2026Are We Really Measuring Forgetting? Pitfalls in Evaluating Machine Unlearning for Multimodal LLMsarXiv:2610.19613v1 Announce Type: new Abstract: As machine unlearning for multimodal large language models (MLLMs) rapidly advances, rigorous evaluation of its effectiveness becomes increasingly impor→15 Oct 2026An Explainable System for Offensive Content Detection in Nastaliq Urdu: Corpus Curation, Transformer Benchmarking, and Word-Level AttributionarXiv:2610.15795v1 Announce Type: cross Abstract: Automated moderation systems are deployed at scale, yet they remain opaque to the users and moderators affected by their decisions, and their coverage→15 Oct 2026AdaptFuse: Training-Free Sequential Preference Inference via Externalized Bayesian ComputationarXiv:2604.03925v2 Announce Type: replace Abstract: Large language models (LLMs) struggle to accumulate evidence across multiple rounds of user interactions, failing to update their beliefs in a manne
TechniqueMultimodal8 recent entries15 Oct 2026Are We Really Measuring Forgetting? Pitfalls in Evaluating Machine Unlearning for Multimodal LLMsarXiv:2610.19613v1 Announce Type: new Abstract: As machine unlearning for multimodal large language models (MLLMs) rapidly advances, rigorous evaluation of its effectiveness becomes increasingly impor→15 Oct 2026Are Multimodal Benchmarks Really Useful? Item-Level Multimodal Benchmark Diagnosis via Structure-Response Co-CalibrationarXiv:2610.14542v1 Announce Type: new Abstract: Multimodal benchmarks are central to evaluating whether models can reason over information from multiple modalities. However, existing literature has re→15 Oct 2026Are Benchmarks Reliable? Toward Structural Diagnosis via Item-Level Capability BoundariesarXiv:2609.33196v2 Announce Type: replace Abstract: Evaluating large language models (LLMs) relies heavily on benchmark scores, yet aggregate metrics can obscure whether individual items serve their i→15 Oct 2026AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned PrefixesarXiv:2610.14487v1 Announce Type: new Abstract: Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their mode→15 Oct 2026AesCode: Aesthetic Code Generation with Decoupled Cross-Modal RewardsarXiv:2610.18860v1 Announce Type: new Abstract: Generating information-rich visual artifacts such as slides, posters, and dashboards as code is appealing because the outputs are structured, editable, →15 Oct 2026ADAPT: Audio-Depth Action Planning with Transition-Aware Rewards for Lights-Out Industrial NavigationarXiv:2610.15956v1 Announce Type: new Abstract: Vision-language-action models for embodied navigation typically rely on RGB observations, limiting their applicability in lights-out industrial environm→15 Oct 2026ActICL: Post-Training for Agentic Multimodal In-Context LearningarXiv:2610.19475v1 Announce Type: cross Abstract: Multimodal in-context learning (ICL) is an important capability of multimodal large language models (MLLMs). Its performance is closely tied to the de→15 Oct 2026Accelerating VLA Inference with Perturbation Informed ExecutionarXiv:2610.19656v1 Announce Type: new Abstract: Action-chunking vision-language-action (VLA) models predict sequences of future robot actions. Under fixed-horizon execution, the robot executes a fixed
TechniqueSafety8 recent entries15 Oct 2026Can agents whose sole duty is to keep humans alive be hijacked to kill? Delayed multi-turn derived-action trap attacks in simulated high-stakes environmentsarXiv:2610.15153v1 Announce Type: new Abstract: We ask whether a tool-using agent whose sole duty is to keep humans alive can be induced, by delayed prompt injection, to kill. The keep-alive duty is i→15 Oct 2026Big-Big-SycoBench: A Benchmark to Detect Sycophancy Through Model Responses to Valid vs. Invalid ArgumentsarXiv:2610.15634v1 Announce Type: new Abstract: Sycophancy in large language models risks amplifying unsupported claims or beliefs across the knowledge ecosystem as these systems increasingly contribu→15 Oct 2026Beyond Static Positional Encoding: Bayesian Time-aware Attention for Non-Stationary Long-term Time Series ForecastingarXiv:2610.19603v1 Announce Type: new Abstract: Transformers have demonstrated strong performance in long-term time series forecasting (LTSF), yet modeling non-stationary temporal dependencies remains→15 Oct 2026Beyond Answer Voting: Fine-Grained Correction of LLM Reasoning through Graph Alignment across Sampled TracesarXiv:2610.18606v1 Announce Type: new Abstract: Large language models increasingly tackle tasks that require multi-step reasoning, where errors in intermediate claims can affect the final answer. Self→15 Oct 2026AuthGateBench: Benchmarking LLMs as Authorization GatearXiv:2610.16426v1 Announce Type: new Abstract: Automated approval systems let coding agents execute high-risk actions with minimal human oversight by delegating the authorization decision to a classi→15 Oct 2026AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned PrefixesarXiv:2610.14487v1 Announce Type: new Abstract: Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their mode→15 Oct 2026Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg EquilibriaarXiv:2511.17937v2 Announce Type: replace Abstract: Alignment faking is a form of strategic deception in AI in which models selectively comply with training objectives when they infer that they are in→15 Oct 2026AgentStripes: An Evidence-Graded Maturity Model for AI AgentsarXiv:2610.18603v1 Announce Type: new Abstract: Benchmarks report what an Artificial Intelligence (AI) agent can do on a task suite, but not whether one released version has been engineered with the c