AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchAutoPDEBench: Benchmarking LLM Auto-Research for Neural PDE Solver Design
arXiv:2609.32245v1 Announce Type: cross Abstract: Partial differential equations (PDEs) are essential for modeling complex physical systems, and neural solvers have recently emerged as powerful data-driven tools for nume…
- arXiv cs.AIresearchDS-VLA: A Dendritic-inspired Vision-Language-Action Model for Robust Action Control
arXiv:2609.32253v1 Announce Type: cross Abstract: Vision-language-action (VLA) models have achieved strong performance in language-conditioned manipulation, yet success under nominal evaluation does not necessarily trans…
- arXiv cs.AIresearchSupporting and Performing Culture from the Inside
arXiv:2609.32281v1 Announce Type: cross Abstract: Anthropology and related disciplines which study culture have often found it useful to consider their epistemological and methodological approaches in terms of emic versu…
- arXiv cs.AIresearchA Solvable Theory of Pre-training Data Poisoning: Regime-Dependent Scaling Exponents
arXiv:2609.32288v1 Announce Type: cross Abstract: Pre-training data poisoning of large language models is usually studied using targeted backdoors and their survival through safety post-training, which leaves open a more…
- arXiv cs.AIresearchAffordance-Conditioned Decision Making: Bridging the Semantic-Spatial Gap in Zero-Shot Cross-Floor Vision-and-Language Navigation
arXiv:2609.32292v2 Announce Type: replace-cross Abstract: Vision-and-language navigation increasingly relies on general-purpose semantic planners, yet translating correct high-level intent into reliable physical executio…
- arXiv cs.AIresearchTRAP: Understanding and Mitigating Privacy Memorization in Language Models
arXiv:2609.32293v1 Announce Type: cross Abstract: Fine-tuning a language model on sensitive records can leave it able to reproduce them. We ask when this memorization arises and how to prevent it without knowing in advan…
- arXiv cs.AIagentsWhat Can a Leaderboard Certify? Compositional Controllability for Fair Evaluation and Training of Biomedical Literature-Review Agents
arXiv:2609.32318v1 Announce Type: cross Abstract: Leaderboards rank long-horizon agents by their final outputs. Yet a higher score alone does not establish whether two systems are comparable or which stage accounts for t…
- arXiv cs.AIresearchPhase Space Attention:A Hairer Lift Circumvents the Single-Layer Induction Obstruction
arXiv:2609.32319v1 Announce Type: cross Abstract: We circumvent the Sanford-Hsu-Telgarsky (SHT) single-layer induction obstruction within a linear, one-step, causal, bilinear, symplectically consistent design class on th…
- arXiv cs.AIresearchSuperposed Inference for Hyperdimensional Computing
arXiv:2609.32320v1 Announce Type: cross Abstract: Hyperdimensional computing (HDC) is attractive for efficient and robust learning, but conventional inference still encodes every query independently, repeatedly paying th…
- arXiv cs.AIresearchActive Feature Acquisition With Incomplete Training Data
arXiv:2609.32325v1 Announce Type: cross Abstract: In many prediction tasks, acquiring all features can be a prohibitively expensive or outright impossible task. Further, in many cases a static subset of features may not…
- arXiv cs.AIresearchProgressive-View On-Policy Distillation for Regional-to-Global Transfer in Multimodal LLMs
arXiv:2609.32333v1 Announce Type: cross Abstract: Regional-to-global distillation uses crop-conditioned guidance to improve full-image understanding. The challenge is to effectively transfer the teacher's crop-based adva…
- arXiv cs.AIresearchAnalog-Friendly Predictive Coding without Activation Derivatives
arXiv:2609.32350v1 Announce Type: cross Abstract: Predictive coding (PC) is a local, energy-based alternative to backpropagation (BP) whose iterative inference dynamics make it attractive for implementation on analog har…
- arXiv cs.AIresearchEyeVQA: Benchmarking Ophthalmic Vision-Language Models from Recognition to Spatial Grounding
arXiv:2609.32352v3 Announce Type: replace-cross Abstract: Vision-language models (VLMs) have shown increasing potential for medical image understanding, yet their capabilities in ophthalmic imaging remain insufficiently…
- arXiv cs.AIagentsBlack-Box Auditing of Epistemic Reliability in Multi-Agent Debate Distillation
arXiv:2609.32361v1 Announce Type: cross Abstract: Debate distillation adapts weaker verifiers using multi-agent debate transcripts to improve their judgement in subsequent debates, but gains on monitored tasks do not est…
- arXiv cs.AIresearchDiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series Forecasting
arXiv:2609.32363v2 Announce Type: replace-cross Abstract: Probabilistic time series forecasting requires modeling and predicting complex and time-varying distributions. Recently, Denoising Diffusion Probabilistic Model (…
- arXiv cs.AIagentsMeasurement Boundaries in LLM Financial Agent Evaluation: Fixed-Tape Execution and Multi-Defect Auditing
arXiv:2609.32379v1 Announce Type: cross Abstract: What controls are needed to interpret execution performance and audit scores in LLM agent evaluations? We study two limits on these interpretations in a financial agent h…
- arXiv cs.AIresearchTimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra
arXiv:2609.32384v2 Announce Type: replace-cross Abstract: Real-world time series are inherently non-stationary, with trends, periodic patterns, and uncertainty evolving over time. While the Fourier domain offers a natura…
- arXiv cs.AIagentsDashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard Analysis
arXiv:2609.32385v2 Announce Type: replace-cross Abstract: Interactive dashboards require users to reveal and connect evidence across stateful interactions. Although graphical user interface (GUI) agents could automate th…
- arXiv cs.AIagentsSkillDRE: Dual-Stage Red-Team Evolution of Agent Skills via Pre-Execution and Runtime Feedback
arXiv:2609.32400v1 Announce Type: cross Abstract: Agent skills package instructions, executable code, and task-specific resources into reusable artifacts that agents can improve using execution feedback. The same mechani…
- arXiv cs.AIagentsShared Worlds, Private Minds: Structured Memory for Long-Form Writing as World Creation
arXiv:2609.32401v1 Announce Type: cross Abstract: LLM agents that write long-form fiction need an explicit memory of the evolving storyworld to keep new events consistent with established facts. Such memory must keep het…
- arXiv cs.AIresearchAutomatic Speech Recognition for the Basa\`{a} Language: A Low-Resource Approach
arXiv:2609.32408v1 Announce Type: cross Abstract: The rapid advancement of Artificial Intelligence (AI) and Natural Language Processing (NLP) has revolutionized the way humans interact with machines. Among the most impac…
- arXiv cs.AIagentsRE-0: Verified Recursive Improvement of Embodied Code-as-Policy Agents through Local On-Policy Distillation
arXiv:2609.32416v1 Announce Type: cross Abstract: Code-as-Policy agents accomplish long-horizon embodied tasks by generating and executing code, yet continually improving them with teachers that are stronger but not glob…
- arXiv cs.AIagentsCyberClear: A Benchmark for LLM Agent Systems on APT Attack Chain Provenance
arXiv:2609.32424v1 Announce Type: cross Abstract: Large language model agents have demonstrated promising capabilities in cybersecurity tasks, yet their ability to reconstruct complete Advanced Persistent Threat attack c…
- arXiv cs.AIresearchRethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It
arXiv:2609.32444v1 Announce Type: cross Abstract: We study training-inference mismatch in reinforcement learning with verifiable rewards (RLVR) for large language models, where rollouts are sampled by an inference engine…
- arXiv cs.AIresearchDRAM: Delta-rule Recurrent Associative Memory for Robot Manipulation Policies
arXiv:2609.32453v2 Announce Type: replace-cross Abstract: Robotic manipulation is inherently history-dependent, yet most pretrained robotic policies condition on only the current observation or a short temporal window. E…
- arXiv cs.AIresearchDe-biasing Skeleton-based Action Recognition with Convex Hull Adaptive Shift
arXiv:2609.32454v1 Announce Type: cross Abstract: Skeleton sequences can represent both individual actions and multi-entity interactions, encompassing human bodies, hands, objects, and robots. Existing approaches to reco…
- arXiv cs.AIresearchSeeing Parts, Reasoning about Worlds: Visual Inference under Partial Observation
arXiv:2609.32456v1 Announce Type: cross Abstract: World modeling under partial observation requires reasoning about the complete worlds that remain compatible with limited visual evidence. Occluded objects and unseen reg…
- arXiv cs.AIresearchAdapting Nonstationary Multi-output Gaussian Processes to Bayesian Optimization
arXiv:2609.32464v1 Announce Type: cross Abstract: Multi-objective Bayesian optimization (MOBO) commonly relies on independent Gaussian processes (GPs) with stationary kernels, limiting its ability to represent nonstation…
- arXiv cs.AIresearchBison: Cross-Dataset Learning for Unseen-Compound Perturbation Prediction
arXiv:2609.32467v1 Announce Type: cross Abstract: Predicting transcriptional responses to unseen compounds is limited by fragmented chemical coverage and heterogeneous experimental platforms and gene panels. To assess mo…
- arXiv cs.AIresearchExplaining Textual Entailment with Lexical Entailments: Using LLMs to Supply Lexical Relations for Formal Proofs
arXiv:2609.32491v1 Announce Type: cross Abstract: Large Language Models (LLMs) are highly capable of natural language reasoning and appear to store a great deal of lexical knowledge, but it is still unclear how much of t…
- arXiv cs.AIresearchSoFT: Soft Targets for Generalizable LLM Fine-Tuning
arXiv:2609.32493v2 Announce Type: replace-cross Abstract: Distillation enables student language models to acquire new capabilities from expert teachers. However, integrating knowledge from multi-teacher, multi-domain dem…
- arXiv cs.AIagentsHearsay: Can an Auditor Trust the Record a Deployed Agent Harness Writes?
arXiv:2609.32495v1 Announce Type: cross Abstract: An agent harness, the code that turns a model into an agent, writes its own record of each run, and that record is all a later reader gets when a run is disputed, investi…
- arXiv cs.AIresearchLearning an Anchored Prompt Space for Continual Adaptation of Large Language Models
arXiv:2609.32499v2 Announce Type: replace-cross Abstract: Continually adapting large language models requires acquiring new knowledge while preserving previously learned capabilities. Although jointly adapting model para…
- arXiv cs.AIresearchRondo: Unsupervised Discovery of Recurring Temporal Structure
arXiv:2609.32500v1 Announce Type: cross Abstract: Many real-world time series data exhibit structural properties at multiple scales, from short, recurring units to complex sequences composed of these units. Unsupervised…
- arXiv cs.AIagentsREFINE: A Resilient Evolution Framework for Intelligent Enterprise Alert Triage in Security Operations Centers
arXiv:2609.32516v1 Announce Type: cross Abstract: Security Operations Centers (SOCs) process large volumes of alerts daily. Alert triage prioritizes high-risk threats while reducing manual review of benign alerts. LLM ag…
- arXiv cs.AIagentsWhen Users Change Their Minds: Measuring and Repairing Intent Drift in LLM Agents
arXiv:2609.32520v1 Announce Type: cross Abstract: LLM agents often operate over multi-turn interactions in which user intent changes before execution. We study intent drift: the failure mode in which superseded parts of…
- arXiv cs.AIresearchActivation Flow: Manufacturing Activations for Steering
arXiv:2609.32530v2 Announce Type: replace-cross Abstract: Difference-in-means steering requires activations recorded while a model shows the desired behavior, which a sandbagging model withholds by deliberately underperf…
- arXiv cs.AIresearchWhat Does a ProcGen Generalization Gap Measure? Action Rules, Residual Entropy, and the Missing Random Floor
arXiv:2609.32532v2 Announce Type: replace-cross Abstract: A generalization gap in reinforcement learning, return on training levels minus return on held-out levels, is usually reported without a reference point. We argue…
- arXiv cs.AIresearchDepthBench: Measuring How Residual Connections Enable More Computational Depth
arXiv:2609.32534v1 Announce Type: cross Abstract: Depth is a natural way to increase the computational capacity in Transformers, yet the contribution of deeper layers can diminish as depth grows larger. Recent approaches…
- arXiv cs.AIagentsDo Audio LLMs Listen Before They Act? Diagnosing Acoustic-Context Gating in Voice Agents
arXiv:2609.32536v1 Announce Type: cross Abstract: Audio language models can recognize spoken commands and invoke tools, but an agent must first decide whether the acoustic and conversational context warrants action. We i…
- arXiv cs.AIresearchIn-Flight KV Cache with Clean Anchors for Faster Autoregressive Video Diffusion
arXiv:2609.32540v2 Announce Type: replace-cross Abstract: Few-step autoregressive video diffusion generates a long video by splitting the video into temporal chunks and generating chunk-by-chunk, each through a short seq…
- arXiv cs.AIresearchHarnessing Coupled Stream Completion For Human-Object Interaction Modeling
arXiv:2609.32551v2 Announce Type: replace-cross Abstract: Text-conditioned human-object interaction (HOI) generation requires body motion, object trajectories & rotations, and hand articulation to remain coordinated. The…
- arXiv cs.AIresearchFeature Space Guidance for Breast Cancer Classification in DCE-MRI
arXiv:2609.32555v1 Announce Type: cross Abstract: Dynamic contrast enhanced breast MRI (DCE-MRI) is a powerful clinical tool for breast cancer detection, providing high resolution anatomical detail together with rich tem…
- arXiv cs.AIagentsRetrieved but Not Delivered: Multimodal Memory Delivery for Long-Term Agents
arXiv:2609.32590v1 Announce Type: cross Abstract: Work on memory for multimodal agents optimizes what is written, updated and retrieved. Between retrieval and the answer, however, is a stage that multimodal memory evalua…
- arXiv cs.AIresearchRECAST: Recasting Vision-Language Semantics into an Actionable Cost Map for Robot Navigation
arXiv:2609.32595v1 Announce Type: cross Abstract: Safe and robust robot navigation across diverse environments requires a high-level understanding of complex scenes and the ability to carry it into stable motion. Recent…
- arXiv cs.AIresearchGAUGE: Group-Wise View-Inconsistency Rectification for Feed-Forward 4D Tracking
arXiv:2609.32596v2 Announce Type: replace-cross Abstract: Feed-forward models regress dense 3D point trajectories directly from monocular video, yet the residual after global alignment is substantial and lacks a structur…
- arXiv cs.AIresearchNot Every Term Adds New Structure: Sobolev Novelty for Symbolic Regression
arXiv:2609.32597v1 Announce Type: cross Abstract: Symbolic regression (SR) aims to discover compact and meaningful mathematical equations from data, but searching the vast combinatorial space of symbolic structures remai…
- arXiv cs.AIresearchAnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation Repulsion
arXiv:2609.32602v1 Announce Type: cross Abstract: Adversarial attacks optimized on a single open-weight LLM can transfer to and jailbreak architecturally different models, allowing an attacker with white-box access to on…
- arXiv cs.AIresearchThe Alignment Paradox: How Post-Training Amplifies Confident Hallucinations in Language Models
arXiv:2609.32617v1 Announce Type: cross Abstract: Large language models (LLMs) can produce factually incorrect answers with high confidence, undermining their reliability and limiting the effectiveness of uncertainty-bas…
- arXiv cs.AIresearchIntuition vectors
arXiv:2609.32619v1 Announce Type: cross Abstract: Large self-supervised vision models learn representations that support scene segmentation and the semantic decomposition of physical objects. We ask whether their represe…