PreprintNot peer reviewedArtificial IntelligencePreprint proposes meta-cognitive LLM framework for budget-aware reasoning across problemsOriginal title: ROI-Reasoning: Rational Optimization for Inference via Pre-Computation Meta-CognitionA preprint presents ROI-Reasoning, a two-stage method letting language models judge problem difficulty and allocate a shared token budget across tasks.Muyang Zhao et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom Microsoft ResearchArtificial IntelligencePreprint proposes bias control and injection to improve LLM multi-agent answer aggregationOriginal title: Bias Control and Injection for Multi-Agent AggregationA preprint describes a training-free framework that normalises reasoning style and then stress-tests answers to pick better final outputs from multiple LLM agents.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceA preprint reports an LLM taught to pick and interpret EHR data for ED riskOriginal title: Incentivizing Tool-Use Reinforcement Learning for Post-Arrival Emergency Care PredictionResearchers report a 4B-parameter LLM trained with tool-use reinforcement learning matches larger models at predicting emergency department outcomes.Pengfei Hu et al. · Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceOracle-based process rewards boost agentic LLM training across three puzzle gamesOriginal title: Verifiable Process Rewards for Agentic ReasoningA preprint reports that turning task-solving oracles into dense, action-level rewards improves agentic reasoning training over outcome-only methods.Huining Yuan et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom BaiduArtificial IntelligenceResearchers report a lightweight framework for agentic reinforcement learning training of LLMsOriginal title: A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated RecipesA preprint describes LoongSage, an open-source agentic RL framework designed to cut overhead and keep training and inference in sync.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers report training LLMs on real user behaviour logs boosts user-task performanceOriginal title: Vela: User Behavior as Supervision Sparks LLM IntelligenceA preprint describes Vela, a framework that turns anonymised user behaviour data into training tasks, reportedly improving both user-related and general LLM performance.Yang Xiao et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom DialpadArtificial IntelligenceResearchers report GPT-5.4 leads new benchmark for LLM customer-support skill executionOriginal title: Evaluating LLMs as Agentic Workflow Executors in Real-World Customer Support SkillsA preprint evaluates 17 LLMs as customer-support workflow executors, finding judge disagreement is a significant factor in results.Added 6 Nov 2026
PreprintNot peer reviewedFrom Style3DArtificial IntelligenceAgentic system converts real robot videos into simulatable digital twinsOriginal title: Agentic Real2Sim: Physics-based World Modeling with Vision-Language AgentsA preprint describes an agent-based pipeline that turns recorded robot interactions into physics simulations for policy testing and training.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceSurvey maps how large language models could assist PDE research workflowsOriginal title: Large language models for partial differential equation workflowsA preprint reviews how LLMs might help discover, solve and optimise partial differential equation problems, and finds current benchmarks remain fragmented.Han Wan et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom XiaomiArtificial IntelligenceResearchers propose Prorsa to speed up agentic RL rollout via adaptive prefill-decode splitOriginal title: Prorsa: Unleashing the Potential of Prefill-Decode Disaggregation in Agentic Reinforcement LearningA preprint presents Prorsa, a framework that forecasts prefill and decode demand to speed up agentic reinforcement learning rollouts.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligencePreprint proposes lightweight sanitiser to protect LLM judges from score-hijacking promptsOriginal title: StableJudge: A Lightweight Adversarial Defense Framework for LLM-as-a-JudgeA preprint introduces StableJudge, a small detachable module that strips injected instructions before an AI judge scores text.Anning Hu et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom AmazonArtificial IntelligenceResearchers report RL-trained agent judges and diagnoses competitive-programming code submissionsOriginal title: Beyond the Verdict: Agentic Code Judging with Verifiable FeedbackA preprint describes Arbiter, an agentic verifier trained with reinforcement learning to judge and diagnose code submissions using checkable counterexamples.Added 6 Nov 2026
PreprintNot peer reviewedFrom JD.com / TencentArtificial IntelligencePreprint proposes attention-aware KV cache compression method called BELTOriginal title: BELT : BEhavior-Aligned Low-Rank KV Cache Compression with Learned Top-k Rank AllocationA preprint reports a KV cache compression scheme that targets attention-output quality instead of raw key reconstruction, claiming strong accuracy and speed gains.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceStudy finds LLMs often justify human-rights violations through reasonable-sounding logicOriginal title: UDHRBench: Benchmarking Human Rights in Large Language ModelsA preprint introduces UDHRBench, a dataset testing whether large language models align with the Universal Declaration of Human Rights.Alak Raad et al. · Added 6 Nov 2026
PreprintNot peer reviewedFrom Luma AIArtificial IntelligenceResearchers report mixed results on whether robot agents truly learn or just retryOriginal title: SEAR: Self-Evolving Agentic RoboticsA preprint finds robotic agents sometimes just resample attempts rather than genuinely learning, with transfer depending on code, not task-specific tricks.Added 6 Nov 2026
PreprintNot peer reviewedFrom logsight.aiArtificial IntelligenceResearchers report lattice-based quantization shrinks LLM key caches without quality lossOriginal title: E8 Lattice Key Quantization for Low-Bit KV-Cache CompressionA preprint describes an E8 lattice vector quantization scheme that compresses LLM key-value caches heavily while preserving near-baseline quality and decode speed.Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligencePreprint reports training-free way to share KV caches between different-sized language modelsOriginal title: Think Before You Speak: Enabling Training-Free Latent Communication Between Language ModelsA preprint describes CacheFuse, a training-free method that blends KV caches from two language models to lift accuracy and cut latency.Rishi Khare · Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers report a stealthy backdoor that leaks private RAG documents via a third-party modelOriginal title: RAGLeak: Supply-Chain Privacy Backdoors in Retrieval-Augmented GenerationA preprint describes a supply-chain backdoor that makes a RAG generator secretly encode retrieved private documents for later recovery.Yuchen Chen et al. · Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceA preprint maps how verifier mistakes shape reinforcement learning for reasoning modelsOriginal title: Checking the Checkers: A Synthesis of Verifier Noise and Robustness in RLVRA survey of 58 studies finds that whether a reward checker's errors can be exploited matters more than how often it errs.Savyasachi Gupta et al. · Added 6 Nov 2026
PreprintNot peer reviewedArtificial IntelligenceA preprint reports a graph-based retrieval method for more secure LLM code generationOriginal title: SEEK: Agentic Knowledge Graph Retrieval for Secure Code GenerationResearchers report that a CWE-structured knowledge graph helps LLMs generate more secure code without retraining.Varun Gadey et al. · Added 6 Nov 2026
