AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIagentsHow To Train Your World Model: Fine-tuning vs RAG for LM-based World Modeling
arXiv:2610.02542v1 Announce Type: new Abstract: World models (WMs) simulate the transition dynamics of environments, enabling agents to plan over the consequences of their actions. In text-based environments, fine-tuning…
- arXiv cs.AIresearchHow to Have a Sensitive Debate: An Instance-Optimal Protocol for AI Debate
arXiv:2610.02557v1 Announce Type: new Abstract: As powerful AI systems reach and sometimes surpass the abilities of human experts across a range of cognitively demanding tasks, the problem of accurate oversight and super…
- arXiv cs.AIresearchMitigating Social Sycophancy via Pluralistic Preference Optimization
arXiv:2610.02568v2 Announce Type: replace Abstract: Personal advice, including relationship advice, now ranks among the most common uses of generative AI. But language models (LMs) exhibit sycophancy: they affirm users m…
- arXiv cs.AIresearchAnswering clinicians' questions over trial evidence tables with verifiable, feedback-driven language models
arXiv:2610.02576v1 Announce Type: new Abstract: Systematic reviews condense clinical trials into evidence tables, yet clinicians can interrogate these tables only through database queries, and many questions concern attr…
- arXiv cs.AIresearchLabels Override Definitions in Jev-Style Typed Decision Models
arXiv:2610.02586v1 Announce Type: new Abstract: A typed decision model answers a fixed question about an input by returning a probability for each of several caller-defined options. Each option carries a short label and…
- arXiv cs.AIagentsOpen-Endedness Bench: Measuring Epistemic Process from Agent Records
arXiv:2610.02588v1 Announce Type: new Abstract: Agents are increasingly given open-ended research tasks: discovering an empirical law from self-designed experiments, improving a heuristic whose optimum nobody knows, or b…
- arXiv cs.AIresearchTasteBench: Multimodal Benchmark for Sensory Prediction, from Molecules to Sustainable Foods
arXiv:2610.02599v1 Announce Type: new Abstract: Sustainable protein discovery lacks the fast computational proxies, analogous to molecular docking or density functional theory, that accelerate drug and materials discover…
- arXiv cs.AIresearchTime Series Forecasting Benchmarks Need Scenario-Grounded Stress Testing
arXiv:2610.02608v1 Announce Type: new Abstract: Time series forecasting (TSF) increasingly drives decisions in transportation, energy, finance, healthcare, and infrastructure, yet current evaluation remains overly narrow…
- arXiv cs.AIagentsVERSE: Verified Self-Evolving Optimizer for Agent Harnesses
arXiv:2610.02616v1 Announce Type: new Abstract: Harness evolution improves an LLM agent's prompts, tools, and workflow, while the optimizer's own tools and procedures often remain fixed. We study whether an optimizer can…
- arXiv cs.AIresearchCuBEs: Culturally-Situated Behavioral Evaluations and the Limitations of Culture-Blind LLM Judges
arXiv:2610.02622v1 Announce Type: new Abstract: Evaluating the occurrence and triggers of large language model (LLM) behaviors - such as sycophancy, self-preference, or over-confidence - is critical for predicting real-w…
- arXiv cs.AIagentsLost in the Request: How Communication Variation Disrupts Retrieval and Action in Email Agents
arXiv:2610.02627v1 Announce Type: new Abstract: An email assistant should not complete less work simply because a user phrases the same request differently. Yet most benchmarks test each task with only one canonical requ…
- arXiv cs.AIresearchDesigning the Future of User Feedback for Generative AI
arXiv:2610.02631v1 Announce Type: new Abstract: Post-deployment feedback from users can be a cost-effective, scalable, and representative means to monitor and improve generative AI systems and features. When implemented…
- arXiv cs.AIagentsBatched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript
arXiv:2610.02638v1 Announce Type: new Abstract: Full-duplex voice agents make many small, closed decisions, which current systems answer by slow autoregressive decoding. We propose DuplexJev, which feeds ASR-encoder hidd…
- arXiv cs.AIagentsCoherence-Driven Belief Formation and Population Dynamics of Contagion in LLM Agents
arXiv:2610.02654v1 Announce Type: new Abstract: Models of social contagion usually assume how individuals adopt beliefs and derive population behavior from it. We instead empirically measure belief adoption in language m…
- arXiv cs.AIagentsA GHOST in Long-Horizon Agents: Governance Hazard from Overlooked Safety Constraints across Turns
arXiv:2610.02664v1 Announce Type: new Abstract: Long-horizon agents are now playing an increasingly significant role in assisting humans with complex problem-solving. However, it is exactly their extended interaction his…
- arXiv cs.AIresearchSpend Teacher Tokens Where They Matter: Success-Referenced On-Policy Distillation
arXiv:2610.02678v1 Announce Type: new Abstract: On-policy distillation (OPD) combines student-generated rollouts with dense token-level supervision from a teacher, but providing such supervision for every rollout require…
- arXiv cs.AIresearchDataWeave: Deploying Human-LLM Analytics for Exploratory Structured Data Analysis
arXiv:2610.02679v1 Announce Type: new Abstract: Data journalism, the practice of using data analysis to surface newsworthy stories, depends increasingly on the ability of reporters and investigative journalists to uncove…
- arXiv cs.AIresearchLarge language models exhibit unreliable updating of clinical judgment as patient evidence evolves
arXiv:2610.02684v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly explored for clinical reasoning, but whether they appropriately revise judgments as patient evidence evolves remains unclear.…
- arXiv cs.AIagentsDecoupling Memory from Context: Structured Memory for Token-Efficient Test-Time Continual Learning
arXiv:2610.02687v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in enterprise, scientific, and medical applications, where agents must incorporate domain-specific knowledge and…
- arXiv cs.AIresearchLearning to Revise Reasoning with Segment-wise On-Policy Distillation
arXiv:2610.02703v1 Announce Type: new Abstract: On-policy distillation (OPD) improves large language model reasoning by training students on their own rollouts with dense token-wise supervision from the teacher. However,…
- arXiv cs.AIresearchLabel-Efficient Time Series Classification at Scale: A Dual-Stream OSSE-LSTM with Counterfactual Attribution
arXiv:2610.02704v1 Announce Type: new Abstract: Time series are produced continuously at enormous scale by industrial equipment, wearables, power grids, and clinical monitors, yet annotation remains manual, expensive, an…
- arXiv cs.AIagentsEgo2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning
arXiv:2610.02715v1 Announce Type: new Abstract: Egocentric videos capture how people carry out everyday activities, yet testing an agent requires evaluating the consequences of actions it chooses itself. We introduce Ego…
- arXiv cs.AIresearchOn the Chain-of-Thought Monitorability of Looped Language Models
arXiv:2610.02741v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring provides a promising approach for detecting undesirable model behavior. Looped language models (LoopLMs) repeatedly apply shared transform…
- arXiv cs.AIresearchDynamic LLM Routers are Often Misguided
arXiv:2610.02762v1 Announce Type: new Abstract: Dynamic LLM routers promise to cut inference costs by sending each query to the cheapest model that can answer it correctly. We analyze six commercial routers across 14 set…
- arXiv cs.AIresearchLaw And Order: Tax Law Autoformalization
arXiv:2610.02792v1 Announce Type: new Abstract: Legal systems are increasingly implemented through software, yet scalable methods for translating legal texts into accurate symbolic representations remain underdeveloped.…
- arXiv cs.AIresearchPAPER2LLM++: Continual Self-Evolution of LLMs from Research Papers
arXiv:2610.02793v2 Announce Type: replace Abstract: Research on LLMs continually uncovers model limitations, their causes, and potential solutions. Yet these human discoveries remain largely disconnected from model evolu…
- arXiv cs.AIresearchModeling Shared and Individual Structure for Cross-Subject Continuous Affect Regression from EEG-fNIRS
arXiv:2610.02796v1 Announce Type: new Abstract: Continuous, second-by-second valence-arousal estimation from physiological signals is typically studied in a subject-dependent setting, where the model sees labeled data fr…
- arXiv cs.AIresearchBitNest: Bit-Nested Speculative Decoding for Memory-Efficient LLM Inference Acceleration
arXiv:2610.02800v2 Announce Type: replace Abstract: Speculative decoding accelerates autoregressive generation by using a lightweight draft to propose multiple tokens for parallel verification. However, existing methods…
- arXiv cs.AIresearchVIGOR: Zero-Shot Visual Generalization via Latent-Space Consistency in Model-Based Reinforcement Learning
arXiv:2610.02801v1 Announce Type: new Abstract: Model-based reinforcement learning (MBRL) achieves strong sample efficiency by planning within learned latent dynamics, yet its performance degrades substantially under uns…
- arXiv cs.AIresearchROUTEAUDIT: Interaction-Aware Identification for Budgeted Multi-Verifier Routing
arXiv:2610.02808v2 Announce Type: replace Abstract: Adaptive multi-verifier systems are commonly compared through endpoint quality-cost gaps, even when the verifier catalog, availability, accounting, information filtrati…
- arXiv cs.AIresearchiS-KV: Online Low-Rank KV Cache Compression via Block-Incremental SVD
arXiv:2610.02815v1 Announce Type: new Abstract: Long chain-of-thought reasoning substantially increases KV-cache memory during autoregressive decoding, as every generated token introduces new key and value states and cau…
- arXiv cs.AIresearchMetaRubric: Learning to Reward for Rubric-Based Reinforcement Learning
arXiv:2610.02824v1 Announce Type: new Abstract: Rubric-based reinforcement learning extends reward-driven optimization to open-ended tasks by assigning partial credit to individual response requirements. However, rubric…
- arXiv cs.AIresearchScaling Trajectories for Complex Tasks through Recursive Self-Rewrite
arXiv:2610.02826v1 Announce Type: new Abstract: Successful trajectories on difficult tasks provide valuable supervision for model improvement, but specialized harnesses introduce interventions that may be unavailable dur…
- arXiv cs.AIresearchMLCommons Jailbreak Benchmark v1.0
arXiv:2610.02827v1 Announce Type: new Abstract: Modern AI systems are designed to refuse hazardous requests. A jailbreak is a prompt crafted to bypass those safeguards and elicit outputs that the system would normally re…
- arXiv cs.AIresearchFSPO: Policy-Consistent Risk and Pareto-Feasible Control for Budgeted LLM RL Post-Training
arXiv:2610.02828v2 Announce Type: replace Abstract: Adaptive LLM reinforcement-learning post-training changes multiple training actuators online, including rollout temperature, group size, clipping, KL regularization, ve…
- arXiv cs.AIresearchAMBER: Multi-View Adaptive Budget Allocation for Listwise Vision-Language Reranking
arXiv:2610.02831v1 Announce Type: new Abstract: Vision-language models (VLMs) are powerful listwise rerankers for multimodal retrieval, but high inference costs restrict them to evaluating small local candidate views. Ex…
- arXiv cs.AIresearchDNAlign: Dynamic Null-Space Safe Alignment for LLMs
arXiv:2610.02844v1 Announce Type: new Abstract: Ensuring the safe and reliable deployment of large language models (LLMs) remains a fundamental challenge. Existing safety alignment approaches either incur high computatio…
- arXiv cs.AIresearchBounded Reachability & Jailbreak Detection via Contraction-Constrained State Space Models
arXiv:2610.02853v1 Announce Type: new Abstract: Safety heads are lightweight classifiers attached to pretrained language models for flagging harmful inputs before generation. Their empirical detection performance has bee…
- arXiv cs.AIagentsHarness-Aware Distillation for Small Language Model Agents
arXiv:2610.02858v1 Announce Type: new Abstract: Language model agents are deployed with a harness, the software around the model that manages its context, tools, and feedback. When such an agent is distilled into a small…
- arXiv cs.AIresearchTACD: Distilling Efficient Text-to-Motion Models via Terminal Amplification Control
arXiv:2610.02867v1 Announce Type: new Abstract: Recent text-to-motion models have improved motion quality and instruction following, yet many-step denoising and large model components make deployment slow and memory-inte…
- arXiv cs.AIagentsPsyEvo: A Personalized Counseling Agent That Self-Evolves at Test Time
arXiv:2610.02885v1 Announce Type: new Abstract: Mental health disorders affect a substantial proportion of the global population, yet a persistent shortage of trained practitioners leaves the majority without adequate ca…
- arXiv cs.AIagentsInterpreting at Write Time: A Policy Ablation for Multi-Goal Agent Memory
arXiv:2610.02897v1 Announce Type: new Abstract: A long-running assistant cannot keep everything it has seen, so it summarises. Summarising is not neutral: what is kept is chosen against some notion of what the record is…
- arXiv cs.AIresearchLUMOS: Tracing Parametric Knowledge from Training Data to Behavioral Outputs in LLMs
arXiv:2610.02902v3 Announce Type: replace Abstract: Current analyses of LLMs' parametric knowledge are largely output-centric, drawing conclusions about what a model knows without verifying what it was actually trained o…
- arXiv cs.AIresearchFrequency Is Not Sensitivity Identifying Safety-Sensitive Experts in Sparse MoE LLM
arXiv:2610.02910v1 Announce Type: new Abstract: Suppressing a small set of routed experts can weaken the safety behavior of a sparse Mixture-of-Experts (MoE) language model without retraining. Which experts to suppress i…
- arXiv cs.AIagentsHASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence
arXiv:2610.02920v1 Announce Type: new Abstract: Agent harnesses play a critical role in defenses by enforcing safety constraints to prevent unsafe actions. However, rapidly emerging attacks outpace manual harness adaptat…
- arXiv cs.AIagentsPositive-Unlabeled Learning for Agent Safety False Alarm Auditing
arXiv:2610.02925v1 Announce Type: new Abstract: Safety monitors help safeguard language-model agents interacting with external tools and environments, but conservative monitoring can generate many false alarms, consuming…
- arXiv cs.AIagentsWhen to Compile a Computer-Use Agent? Measuring Payback and Making Compilation Decisions for Token Efficiency
arXiv:2610.02932v1 Announce Type: new Abstract: Compiling GUI procedures that agents execute repeatedly into programs can reduce their token costs. However, measuring payback and deciding when to compile have two challen…
- arXiv cs.AIagentsContinual Graph Memory for Mathematical Research Agents
arXiv:2610.02945v1 Announce Type: new Abstract: Using frontier agent harnesses to tackle mathematical research problems has emerged as an effective means of advancing mathematics. However, solving frontier problems in ma…
- arXiv cs.AIresearchReasoning with Evidence, Not Merely Rationales: Verifiable Preference Proofs for LLM-Based Recommendation
arXiv:2610.02968v1 Announce Type: new Abstract: Large language models (LLMs) can infer user preferences from interaction histories and reviews, yet the rationales they generate may not reflect the information actually us…
- arXiv cs.AIresearchCreateScore: Domain-Theory-Informed Bayesian Routing for LLM-Based CV Screening
arXiv:2610.02972v1 Announce Type: new Abstract: Large language models (LLMs) can support rubric-based screening of CVs, but applying a high-capability model to every candidate and criterion is costly. We present CreateSc…