AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchEncoded but Exposed: Language Embeddings can Reveal More Than They Should in Recommender Systems
arXiv:2610.21713v1 Announce Type: cross Abstract: Language embeddings are increasingly used in recommenders to represent item content, but their privacy risks remain poorly understood. We study whether item embeddings ca…
- arXiv cs.AIresearchFIRM: Fisher-Informed Retention Merging for Federated LLM Unlearning
arXiv:2610.21720v1 Announce Type: cross Abstract: Federated LLM unlearning addresses targeted knowledge removal when forget and retain data remain distributed across private clients and are inaccessible to the server. Th…
- arXiv cs.AIresearchRetrieved but Not Read: Understanding Why Search-Augmented LLMs Fail to Use Available Evidence
arXiv:2610.21731v1 Announce Type: cross Abstract: Web search allows LLMs to access up-to-date information, but it remains unclear why they sometimes produce outdated answers even when relevant evidence is available. To s…
- arXiv cs.AIagentsReliable Knowledge Acquisition for Continual Learning Agents
arXiv:2610.21739v1 Announce Type: cross Abstract: As agents powered by foundation models move into open-ended environments, continual learning from experience becomes essential. For agents built on models with fixed para…
- arXiv cs.AIagentsThe Delegation Dilemma: Economies of Scale, Accountability, and Polycentric Governance
arXiv:2610.21746v1 Announce Type: cross Abstract: Many collective-action systems are neither fully decentralized nor fully centralized: agents voluntarily organize into institutions that share organizational costs and de…
- arXiv cs.AIresearchHarmful Partial Interventions and Benign Leakage in Concept-based Models
arXiv:2610.21750v1 Announce Type: cross Abstract: Artificial intelligence systems deployed in high-risk settings increasingly require models that support meaningful human oversight. Concept-based models (CMs) offer a pro…
- arXiv cs.AIresearchCF-AVSS: A Coarse-to-Fine Audio-Visual Speech Separation Framework with Target-Speaker Refinement and Post-Fusion for Real-World AVSE Challenge
arXiv:2610.21773v1 Announce Type: cross Abstract: Audio-visual speech enhancement (AVSE) leverages visual cues to recover target speech from mixtures corrupted by reverberation, noise, and overlapping speakers. We propos…
- arXiv cs.AIresearchDINO-FIL: Multi-Layer Transformer Fusion for Generative Inpainting Forgery Localization
arXiv:2610.21780v1 Announce Type: cross Abstract: We propose DINO-FIL, a transformer-based framework for pixel-level localization of generative image forgeries. The method leverages multi-layer representations from a pre…
- arXiv cs.AIresearchIn the Guise of Correction: How Framing and Cultural Adaptation Erode LLM Safety on Real-World Criminal Cases
arXiv:2610.21789v1 Announce Type: cross Abstract: Existing safety benchmarks rely largely on synthetic prompts with loosely grounded taxonomies and offer little focused analysis of severe crime categories. We introduce G…
- arXiv cs.AIresearchCross-Embodiment Causal Transfer from Human Videos to Robot Manipulation
arXiv:2610.21798v1 Announce Type: cross Abstract: Scaling robot manipulation models requires large amounts of interaction data, yet collecting diverse real-robot experience remains costly. Human manipulation videos, in c…
- arXiv cs.AIresearchProvably Efficient Risk-Sensitive Offline Reinforcement Learning via Optimized Certainty Equivalents
arXiv:2610.21870v1 Announce Type: cross Abstract: Risk-sensitive reinforcement learning (RL) addresses decision-making under uncertainty and adverse outcomes. Learning risk-sensitive policies from fixed datasets is impor…
- arXiv cs.AIresearchAnvil: Scaling Embodied Spatial Intelligence through Knowledge Recirculation
arXiv:2610.21882v1 Announce Type: cross Abstract: Embodied spatial intelligence requires more than visual recognition: models must localize entities, reason about their relations, and express this knowledge in action-rel…
- arXiv cs.AIresearchIn-Context Learning Before Specialisation: Theory of Learning in a Solvable Topic Model
arXiv:2610.21901v1 Announce Type: cross Abstract: Sequence models can generalise in fundamentally distinct ways: extrapolating structures seen in training, or using their prompt to adapt to novel tasks. The prevailing st…
- arXiv cs.AIagentsLearning Rewards in Strategic Games
arXiv:2610.21912v1 Announce Type: cross Abstract: Understanding the objectives that drive agents' behavior is crucial in strategic environments, where multiple agents interact and adapt to one another. This challenge ari…
- arXiv cs.AIagentsCodeHealth Predicts the Token Cost of Agentic Code Review: A Pre-Registered Study of 465 Agent-Authored Pull Requests
arXiv:2610.21914v1 Announce Type: cross Abstract: Context: Organizations increasingly delegate code review to multi-agent large language model (LLM) pipelines billed per token. Low code quality raises human maintenance e…
- arXiv cs.AIresearchBridgeAfford: Bridging Query-Target Gap for Generalizable 3D Affordance Grounding
arXiv:2610.21918v1 Announce Type: cross Abstract: Language-guided 3D affordance grounding aims to localize functional regions on 3D objects according to interaction intents. However, existing methods exhibit poor general…
- arXiv cs.AIagentsAgentic-Priors: Using Large Language Models to Address the Cold Start Problem in Preference Elicitation
arXiv:2610.21938v1 Announce Type: cross Abstract: An agent that learns a user's preferences from their choices faces a cold start problem: before any choices are observed, it must rely solely on a prior. Informative prio…
- arXiv cs.AIresearchFollowing Delegation Traces to Study Human Oversight in Interaction with Multimodal AI
arXiv:2610.21948v1 Announce Type: cross Abstract: As multimodal AI systems are becoming our everyday thought partners, the shift towards studying how users transfer control to these systems becomes especially urgent. In…
- arXiv cs.AIresearchWhy Shallow Neural Networks Learn Sinusoidal Components First: An Orthogonal Polynomial Dynamics Framework
arXiv:2610.21964v1 Announce Type: cross Abstract: Neural networks often learn low-frequency components before high-frequency ones --- a phenomenon called spectral bias or termed F-principle. Many numerical studies have d…
- arXiv cs.AIresearchTime to Fake It: Continual Machine Learning Intensifies Unfairness in Deepfake Detection
arXiv:2610.21974v1 Announce Type: cross Abstract: Deepfake detectors safeguard against misinformation, rendering their failures as much a societal as a technical concern. As generators constantly evolve, detectors traine…
- arXiv cs.AIresearchGraphBEAT: Graph-based Bucketized Event Aggregation Over Time for Early Risk Stratification of Gastrointestinal Cancers
arXiv:2610.22014v1 Announce Type: cross Abstract: Longitudinal electronic health records (EHRs) contain complementary diagnostic and laboratory information about patient health. However, irregularity, sparsity, heterogen…
- arXiv cs.AIagentsPriming Base Models for Agentic Post-Training: An Open Midtraining Recipe
arXiv:2610.22019v1 Announce Type: cross Abstract: Frontier models, general-purpose and domain-specific alike, now include a midtraining stage between pretraining and post-training, but little has been published about how…
- arXiv cs.AIresearchDemoStream: Batched Task-and-Motion-Plan Optimization for Automated Learning from Demonstration
arXiv:2610.22020v1 Announce Type: cross Abstract: Learning from demonstration is an effective strategy for endowing robots with manipulation capabilities but requires training on datasets comprised of massive amounts of…
- arXiv cs.AIagentsSelection Is Not Learning: Choosing Model Futures
arXiv:2610.22029v1 Announce Type: cross Abstract: Self-evolving language models train on tasks they generate, so every round must decide which updated model the next round inherits. Current loops such as R-Zero, Agent0,…
- arXiv cs.AIresearchMomentum, Commentary Tone, and Earnings Surprise in LLM Financial Forecasts
arXiv:2610.22043v1 Announce Type: cross Abstract: Which inputs drive a language model's financial forecast? We construct synthetic earnings announcements that independently reverse 20-day momentum, non-informational comm…
- arXiv cs.AIresearchDecoding Naturalistic Speech from fMRI with Temporal Modeling and Multimodal Alignment
arXiv:2610.22049v1 Announce Type: cross Abstract: Reconstructing what a person perceives from their brain activity is a stringent test of whether neural recordings carry meaningful information. Naturalistic speech offers…
- arXiv cs.AIresearchMemory as Weights: Internalizing Long-Term History for Streaming Videos
arXiv:2610.22070v1 Announce Type: cross Abstract: Streaming vision-language models (VLMs) must continuously integrate information from the past while remaining responsive to current observations and user queries. Existin…
- arXiv cs.AIresearchMulti-SFT: Permissively-Licensed Multilingual Instruction-Following
arXiv:2610.22086v1 Announce Type: cross Abstract: Supervised fine-tuning datasets remain overwhelmingly English-centric and commercially restricted, as their content is frequently generated by proprietary models whose te…
- arXiv cs.AIresearchTIPS: A Tiny Inference-time Policy for Scaling in Diffusion Language Models
arXiv:2610.22099v1 Announce Type: cross Abstract: Masked diffusion language models expose an inference-time choice: which positions become visible after each denoising pass. Many existing decoding methods use hand-design…
- arXiv cs.AIresearchCausal Integrated Gradients: Scalable Feature Attribution Along Causal Paths
arXiv:2610.22126v1 Announce Type: cross Abstract: Existing feature attribution methods that account for feature dependencies do not scale to high dimensions. We propose causal integrated gradients (CIG) and causal expect…
- arXiv cs.AIresearchExplaining the Roles of Experts in Mixture-of-Experts LLMs for Personalization
arXiv:2610.22127v1 Announce Type: cross Abstract: Personalization of Large Language Models (LLMs) is increasingly important for user-centric applications, yet prior methods mainly adapt dense LLMs, whose polysemantic neu…
- arXiv cs.AIagentsSpecification Gaming and a Constrained Reinforcement Learning Mitigation in a Cleaning Gridworld
arXiv:2610.22152v1 Announce Type: cross Abstract: Reinforcement learning agents optimize exactly the reward signal they are given, which can diverge sharply from an intended true objective when that signal is only a prox…
- arXiv cs.AIresearchEMO: A Closed-Loop Test-Time Computation Framework via Alternating Context Optimization and Logits Steering
arXiv:2610.22160v1 Announce Type: cross Abstract: Test-time context optimization adapts frozen language models through textual feedback, but ordinary conditional generation provides no separate control over how strongly…
- arXiv cs.AIagentsMosaic: Multi-Session Agentic Reinforcement Learning
arXiv:2610.22174v1 Announce Type: cross Abstract: Most existing agentic reinforcement learning methods assume single-session trajectories, where interaction proceeds within one continuously accumulated context. However,…
- arXiv cs.AIresearchMotion- and Gravity-Aware Feedforward 3D Reconstruction
arXiv:2610.22177v1 Announce Type: cross Abstract: Feed-forward transformer models have recently achieved strong performance in multi-view 3D reconstruction, but they still rely primarily on appearance-based attention and…
- arXiv cs.AIresearchWhole-Body Dexterity
arXiv:2610.22184v1 Announce Type: cross Abstract: Current humanoid dexterous manipulation remains largely confined to end-effectors and simple grasps, falling short of humans' ability to engage the entire body by bracing…
- arXiv cs.AIresearchChange the Number, Change the Answer: Assessing Rating Scale Sensitivity in LLM-Based Text Scoring
arXiv:2610.22187v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used in computational social science to assign scalar scores to social and psychological constructs. However, studies use di…
- arXiv cs.AIagentsDecentralized LLM Workflows via Sequential Auctions
arXiv:2610.22201v1 Announce Type: cross Abstract: We present a scalable framework for solving multi-step tasks through a decentralized market of self-interested agents. Prior theoretical work demonstrated that agents org…
- arXiv cs.AIagentsBeyond Downstream QA: Rethinking How We Evaluate Agent Memory
arXiv:2610.22209v1 Announce Type: cross Abstract: Agent memory benchmarks evaluate whether an agent's responses reflect correct knowledge, measuring downstream question-answering accuracy as a proxy for memory quality. W…
- arXiv cs.AIresearchTRACE: Topology-Aware Reconnection via Prediction-Adaptive Connectivity Enforcement for Thin-Structure Segmentation
arXiv:2610.22210v1 Announce Type: cross Abstract: Thin, elongated structures are particularly sensitive to connectivity errors, as even a few local segmentation mistakes can disrupt otherwise continuous paths. Existing s…
- arXiv cs.AIresearchSkill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models
arXiv:2610.22217v1 Announce Type: cross Abstract: Reinforcement learning (RL) can elicit specialized visual capabilities in vision-language models (VLMs), yet jointly optimizing heterogeneous objectives often introduces…
- arXiv cs.AIagentsOS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents
arXiv:2610.22234v1 Announce Type: cross Abstract: Real desktop automation is neither purely visual nor purely programmatic. Agents must understand user-facing interfaces while efficiently manipulating files, scripts, and…
- arXiv cs.AIresearchTraining-Free On-Policy Distillation
arXiv:2610.22236v2 Announce Type: replace-cross Abstract: On-policy distillation consolidates specialist models into one student at the price of rollouts, teacher passes and gradient steps for every teacher. We show that…
- arXiv cs.AIagentsLanguage Is a Security Surface for Agent Guards
arXiv:2610.22244v1 Announce Type: cross Abstract: Step-level guards decide, before each tool call, whether an agent's next action is safe given the user's request and the trajectory so far. Users write those requests in…
- arXiv cs.AIagentsLibrarian Agents: Training LLMs to Curate Long-Context Workspaces that Anticipate Future Queries
arXiv:2610.22273v1 Announce Type: cross Abstract: Reasoning over long document contexts remains a challenge for large language models (LLMs). Standard retrieval methods are inherently reactive, bottlenecking agents with…
- arXiv cs.AIresearchAsking Smarter Questions: Targeted Active Preference Learning for Underspecified Rewards
arXiv:2610.22285v1 Announce Type: cross Abstract: Reward learning typically assumes the human's input covers everything that matters for the task. In practice, humans can only attend to so much at once, so their teaching…
- arXiv cs.AIresearchTowards Continual Learning: Mitigating Forgetting With Subspace Tuning
arXiv:2610.22293v1 Announce Type: cross Abstract: Large language models (LLMs) acquire new skills through fine-tuning, but it can destroy capabilities they already have: catastrophic forgetting remains a central obstacle…
- arXiv cs.AIresearchAI Consilium: a Benchmark of LLM Clinical Multidisciplinary Second Opinions
arXiv:2610.22298v1 Announce Type: cross Abstract: Large language models (LLMs) can generate differential diagnoses for complex cases, but a useful second opinion must also be factually grounded, clinically coherent, acti…
- arXiv cs.AIresearchKnowledge That Is Linear Before Fine-Tuning Generalizes After It
arXiv:2610.22300v2 Announce Type: replace-cross Abstract: Not all facts a language model can recall support further learning equally well. We distinguish active facts, which support generalization beyond their own recall…
- arXiv cs.AIagentsLearning to Steer LLM Agents and Multi-Agent Systems
arXiv:2610.22316v1 Announce Type: cross Abstract: Large language model (LLM)-based agents and multi-agent systems increasingly tackle complex tasks through long-horizon tool use and inter-agent collaboration. However, th…