AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchNullify: Null-Space Activation Steering for Training-Free LLM Unlearning
arXiv:2610.10655v1 Announce Type: cross Abstract: Large Language Models (LLMs) inevitably internalize substantial amounts of sensitive or private information during pre-training, while LLM unlearning aims to selectively…
- arXiv cs.AIresearchBeyond Owls: Subliminal Learning Can Transfer Learned Capabilities and Backdoors
arXiv:2610.10657v1 Announce Type: cross Abstract: In subliminal learning (SL), a teacher model passes on a trait to a student model by distillation on data semantically unrelated to the trait. So far, SL has been demonst…
- arXiv cs.AIresearchLinSlot: Exploiting Linear Representation hypothesis for unsupervised attribute discovery from slot based object representation
arXiv:2610.10722v2 Announce Type: replace-cross Abstract: This paper studies the problem of learning disentangled representations of objects and their attributes from raw, unstructured image data. Slot-based methods have…
- arXiv cs.AIresearchConversational Task Disambiguation over Tabular Data: Leakage-Aware Formulation, Benchmark Suite, and Training
arXiv:2610.10740v1 Announce Type: cross Abstract: Conversational task disambiguation over tabular data uses dialogue to resolve missing information about a user's intended task before producing a solution over tables or…
- arXiv cs.AIresearchBRANCH: Bypassing Multi-Scanner AI Guardrails
arXiv:2610.10742v1 Announce Type: cross Abstract: AI systems increasingly rely on Large Language Models (LLMs) as core reasoning engines, making them targets for prompt injection and jailbreaks. Guardrails monitor and va…
- arXiv cs.AIresearchClarify, Then Focus: Statement Normalization for Conversation Analytics at Scale
arXiv:2610.10758v1 Announce Type: cross Abstract: Enterprise conversation analytics asks many questions of millions of interactions. Each question can require reconstructing what people mean and identifying which informa…
- arXiv cs.AIresearchAI-Mediated Self: How HCI Defines and Relates to the Self
arXiv:2610.10770v1 Announce Type: cross Abstract: How might AI alter how we understand and experience the self? This scoping review analyzes 102 papers to examine how the self is defined in the field of human-computer in…
- arXiv cs.AIagentsMemoWM: How World Models Change What Agents Need to Remember
arXiv:2610.10778v1 Announce Type: cross Abstract: Long-term agents face growing storage demands as they accumulate experience. World models capture reusable regularities that can reduce the information stored for each ex…
- arXiv cs.AIresearchVICO: Visual Environments Co-Evolving for Vision-Language Model Reasoning
arXiv:2610.10782v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a standard recipe for post-training vision-language models (VLMs), but it typically assumes a static trai…
- arXiv cs.AIagentsNavGPT-3: Harnessing Context in a Hierarchical Navigation Runtime
arXiv:2610.10787v2 Announce Type: replace-cross Abstract: Language models trained with long-horizon agentic reinforcement learning can generalize knowledge through reasoning, express precise actions, and pursue goals ove…
- arXiv cs.AIresearchGrammar Concept Annotation at Scale: Deployed Fine-Tuned Small Language Models Outperform Prompted Frontier Models
arXiv:2610.10827v2 Announce Type: replace-cross Abstract: Corrective feedback is among the best-evidenced drivers of second-language acquisition, yet corrections delivered during lessons rarely accumulate into an actiona…
- arXiv cs.AIresearchReal Long-Term Memory for AI: A 50-Million-Token Window That Is Faster and Cheaper Than Recompute
arXiv:2610.10845v1 Announce Type: cross Abstract: A large language model can only use the text that fits in its context window, and it recomputes its internal key-value (KV) state for a prompt every time the prompt is se…
- arXiv cs.AIagentsRFChipAgent: Multi-Agentic AI Flow for Analog/RF Chip Design
arXiv:2610.10858v1 Announce Type: cross Abstract: Analog/RF circuits remain the critical interface between digital computation and the physical world, and emerging standards from Wi-Fi 7 to 6G place stringent demands on…
- arXiv cs.AIagentsStochastic Teacher Intervention for Agentic On-Policy Distillation
arXiv:2610.10878v1 Announce Type: cross Abstract: On-policy distillation (OPD) efficiently transfers capabilities from a stronger teacher to a student language model through dense token-level supervision on student-gener…
- arXiv cs.AIresearchThe Missing Fourth Term for the Emulation Tensor Memory Equilibrium (TME) Model: The Residue Deconstruction Cost
arXiv:2610.10924v1 Announce Type: cross Abstract: The Tensor-Memory Equilibrium (TME) model of "FP8 is All You Need (Part 1)" calculates the execution time of Ozaki Scheme II emulation of fp64 as the maximum of a tensor-…
- arXiv cs.AIresearchSpeedbumps: Rejection Attacks on Speculative Decoding
arXiv:2610.10929v1 Announce Type: cross Abstract: Speculative decoding is a popular technique for increasing the speed and reducing the costs of large language model (LLM) inference by verifying multiple draft tokens in…
- arXiv cs.AIresearchRethinking the Tradeoff Between Temporal Encoding and Nonlinear Computation in Spiking Language Models
arXiv:2610.10933v1 Announce Type: cross Abstract: Spiking language models face a tradeoff between representing continuous semantic features over short temporal windows and retaining costly nonlinear attention operations.…
- arXiv cs.AIresearchTRACE: A Governance Framework for Measuring Explainability Debt in Production AI Systems
arXiv:2610.10957v1 Announce Type: cross Abstract: Production AI systems deployed in high-stakes domains accumulate a governance liability that existing monitoring frameworks fail to detect: the progressive inability to e…
- arXiv cs.AIagentsCross-Provider Review as a Runtime Contract for Coding Agents: A Controlled Pilot and Fault-Injection Study
arXiv:2610.10961v1 Announce Type: cross Abstract: Coding agents increasingly share a workstation while drawing on separate providers and subscription allowances. A second agent can inspect a completed answer, but the cal…
- arXiv cs.AIagentsiAm.md: Robot Skill Self-Assessment through Agentic Introspection for Unknown Open-Vocabulary Domains
arXiv:2610.10962v1 Announce Type: cross Abstract: Agentic AI based on Large Language Model generalization capabilities offers a wide range of potential applications, including planning for embodied tasks. For example, em…
- arXiv cs.AIresearchWhen Citations Mislead? A Claim-Level Benchmark for Legal Hallucination Detection
arXiv:2610.10971v1 Announce Type: cross Abstract: Large language models are increasingly used in legal research and drafting, but they can still produce claims that sound convincing without being supported by the cited s…
- arXiv cs.AIresearchBudgeted Multi-Source Counterfactual Annotation for Off-Policy Evaluation
arXiv:2610.10974v1 Announce Type: cross Abstract: Off-policy evaluation (OPE) estimates the value of a target policy from logged data, but limited behavior-policy coverage can force high-variance reweighting or reward-mo…
- arXiv cs.AIresearchOptimizing Large Language Models with Chained LMOs
arXiv:2610.10975v1 Announce Type: cross Abstract: Muon has motivated a growing family of optimizers that compose multiple matrix normalizations, but these methods remain fragmented and lack a unified perspective. We intr…
- arXiv cs.AIresearchProbabilistic Sensing, Deterministic Authority: Admitting Model-Produced Observations into Sufficiency-Checked Governance Contracts
arXiv:2610.10978v1 Announce Type: cross Abstract: When a field that an authority contract needs exists only in unstructured evidence, a model can sense it. We admit the model's output only as an observation record with a…
- arXiv cs.AIagentsWhy LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation
arXiv:2610.11008v1 Announce Type: cross Abstract: Language-model agents favor their own group because they have watched their members favor each other. The group label alone does little once the decision has a cost; what…
- arXiv cs.AIresearchMid-Training Language Models on Raw Video
arXiv:2610.11019v1 Announce Type: cross Abstract: Multimodal large language models learn mostly from paired image-text data or annotated video, and raw web video is rarely used to further train an existing language model…
- arXiv cs.AIresearchIntent Graph: Navigating the Analytical Reasoning Space for Exploratory Data Analysis
arXiv:2610.11025v1 Announce Type: cross Abstract: Exploratory data analysis (EDA) is rarely open-ended in practice: analysts work from high-level domain questions toward the concrete analyses that can answer them, priori…
- arXiv cs.AIagentsNOMOS: Compiling Written Policies into Statically Verified Tool-Call Gates for LLM Agents
arXiv:2610.11030v2 Announce Type: replace-cross Abstract: Tool-using LLM agents violate the policies they are deployed to enforce, often silently. Prior defenses hand-write rules, query an LLM verifier per action, or com…
- arXiv cs.AIresearchLanguage Modeling is Monotone Compression
arXiv:2610.11031v1 Announce Type: cross Abstract: A long-standing hypothesis in artificial intelligence and neuroscience posits that intelligence is closely related to compression: the ability to compress information eff…
- arXiv cs.AIagentsAffordDrive3D: Affordance-Aware World-Action Modeling with Spatial Understanding
arXiv:2610.11060v2 Announce Type: replace-cross Abstract: World-action models have recently improved autonomous driving by jointly learning future scene prediction and trajectory generation. Most existing approaches mode…
- arXiv cs.AIresearchEmergent Inverse-Depth Scaling From Nonlinearity In Attention
arXiv:2610.11063v1 Announce Type: cross Abstract: Scaling laws describe power-law improvements in model performance with dataset size and parameter count, yet their underlying mechanisms are not fully understood. To expl…
- arXiv cs.AIresearchStability-Plasticity Balance via Singular-Vector Selection in LLM Continual Learning
arXiv:2610.11076v1 Announce Type: cross Abstract: Domain-specific continual adaptation of LLMs risks catastrophic forgetting, creating a fundamental tension between acquiring new capabilities and preserving those learned…
- arXiv cs.AIresearchContinuous Ground-Truth Construction and a Recovery Policy for Air--Water Robotic Tracking
arXiv:2610.11096v2 Announce Type: replace-cross Abstract: Visual tracking across the air-water interface is challenged by splashes, bubbles, refraction, reflections, and abrupt appearance changes that can temporarily inv…
- arXiv cs.AIresearchSFT-as-Context Mitigates Forgetting in Supervised Fine-Tuning
arXiv:2610.11132v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) equips large language models (LLMs) with specialized capabilities, but often comes at the cost of forgetting the general capabilities of thei…
- arXiv cs.AIagentsActiveMedAgent: Cost-Aware Trajectory Learning for Multimodal Medical Diagnosis
arXiv:2610.11140v1 Announce Type: cross Abstract: Clinical diagnosis is inherently sequential: clinicians escalate from cheap to costly tests only when additional evidence is expected to resolve diagnostic uncertainty. W…
- arXiv cs.AIresearchImproving Image-Based Nutrition Estimation Through Multimodal Food-Item Verification and Recovery
arXiv:2610.11144v1 Announce Type: cross Abstract: Single-image nutrition estimation can fail silently when visible foods are missed. Even when a food is correctly identified, its proposed region may not support portion e…
- arXiv cs.AIresearchSP-DocReader: Difference-Aware Self-Play for Precise Document OCR
arXiv:2610.11148v1 Announce Type: cross Abstract: Accurate page transcription remains difficult for vision language models under limited input and training budgets. We present SP-DocReader, a self-play framework for opti…
- arXiv cs.AIresearchCARE: Constrained Attention Refinement for Fine-Grained Visual Classification via Teacher-Student Distillation
arXiv:2610.11153v1 Announce Type: cross Abstract: Fine-grained visual classification requires models to recognize subtle local traits while exposing the visual evidence behind their predictions. Class-specific attention…
- arXiv cs.AIresearchCharacterizing Statistical Separability in TP-CRIV for Probabilistic AI Models
arXiv:2610.11163v2 Announce Type: replace-cross Abstract: Third-party challenge-response identity verification (TP-CRIV) enables an independent verifier to assess whether a claimant possesses a model identical to a remot…
- arXiv cs.AIresearchPIVOT: Perplexity-Informed KD-to-RL Transition Scheduling for Vertical-Domain Few-Shot Distillation
arXiv:2610.11167v1 Announce Type: cross Abstract: Vertical-domain few-shot classification remains challenging for small language models, as limited supervision makes it difficult to acquire domain-specific decision knowl…
- arXiv cs.AIresearchPMTRM: Pseudo-Memory Temporal Re-encoding Module for Embodied Policy Learning
arXiv:2610.11168v1 Announce Type: cross Abstract: Robotic manipulation often contains repeated motions whose local observations look similar at different phases. When these phases require different actions, a policy that…
- arXiv cs.AIagentsVAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding
arXiv:2610.11171v1 Announce Type: cross Abstract: Long-form video understanding often involves multiple questions about different aspects of the same recording. Yet existing video agents typically process each question t…
- arXiv cs.AIresearchHigher-Order Action Supervision Makes A Strong Policy Class
arXiv:2610.11175v2 Announce Type: replace-cross Abstract: Modern data-driven decision-making methods, such as imitation learning (IL) and reinforcement learning (RL), have achieved great success in solving many complex t…
- arXiv cs.AIresearchRAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation
arXiv:2610.11183v1 Announce Type: cross Abstract: Following retrieved evidence does not guarantee factual correctness: misleading evidence can induce a model to replace an answer it previously gave correctly. Standard ac…
- arXiv cs.AIresearchSafeInferCom: Safe Inference-Time Compute via Verifier-Guided Mid-Generation Intervention for Robotic Task Planning
arXiv:2610.11223v1 Announce Type: cross Abstract: Large Reasoning Language Models (LRLMs) enable multi-step reasoning for robotic task planning, but continued reasoning can overwrite valid intermediate plans or leave con…
- arXiv cs.AIresearchNeuro-Memory Fuzzy Inference System for Mimicking Human-like Car Following Behavior
arXiv:2610.11252v1 Announce Type: cross Abstract: This study presents the Neuro-Memory Fuzzy Inference System (NeMeFIS), a hierarchical machine learning architecture that asymmetrically models acceleration and decelerati…
- arXiv cs.AIresearchGated Memory: Admission-Controlled Memory Formation for Conversational AI
arXiv:2610.11270v1 Announce Type: cross Abstract: Personalized conversational AI relies on long-term memory systems that extract facts from user utterances and store them in persistent vector stores. Despite progress in…
- arXiv cs.AIresearchHow to post-train on a surrogate: Envelope sampling mitigates reward hacking
arXiv:2610.11281v1 Announce Type: cross Abstract: Large language models (LLMs) are commonly post-trained against LLM judges and other cheap surrogates because the true reward, such as human preference, is too expensive t…
- arXiv cs.AIresearchDynaTE: Accelerating Diffusion LLMs via Dynamic Token Execution
arXiv:2610.11284v1 Announce Type: cross Abstract: Diffusion-based LLMs (dLLMs) have recently emerged as a promising alternative to autoregressive (AR) LLMs by enabling bidirectional parallel refinement, alleviating the s…
- arXiv cs.AIagentsREMORY: Learning Residual Memory for Context Compaction
arXiv:2610.11287v3 Announce Type: replace-cross Abstract: Long-horizon agents compact their history to continue within a finite context window, but a textual summary alone may not support every subsequent decision. We in…