AI news archive.
19,861 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchVERGE: Formal Refinement and Guidance Engine for Verifiable LLM Reasoning
arXiv:2601.20055v3 Announce Type: replace-cross Abstract: Despite the syntactic fluency of Large Language Models (LLMs), ensuring their logical correctness in high-stakes domains remains a fundamental challenge. We prese…
- arXiv cs.AIresearchScalable Power Sampling: Unlocking Efficient, Training-Free Reasoning for LLMs via Distribution Sharpening
arXiv:2601.21590v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) post-training is a dominant approach for improving the reasoning performance of large language models (LLMs), yet growing evidence sug…
- arXiv cs.AIresearchOpenVTON-Bench: A Large-Scale High-Resolution Benchmark for Controllable Virtual Try-On Evaluation
arXiv:2601.22725v5 Announce Type: replace-cross Abstract: Recent advances in diffusion models have significantly elevated the visual fidelity of Virtual Try-On (VTON) systems, yet reliable evaluation remains a persistent…
- arXiv cs.AIresearchFrom Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs
arXiv:2602.00628v3 Announce Type: replace-cross Abstract: We investigate the extent to which an LLM's hidden-state geometry can be recovered from its behavior in psycholinguistic experiments. Across eight instruction-tun…
- arXiv cs.AIresearchDCD: Decomposition-based Causal Discovery from Autocorrelated and Non-Stationary Temporal Data
arXiv:2602.01433v3 Announce Type: replace-cross Abstract: Multivariate time series in domains such as finance, climate science, and healthcare often exhibit long-term trends, seasonal patterns, and short-term fluctuation…
- arXiv cs.AIresearchTTT-Parkour: Rapid Test-Time Training for Perceptive Robot Parkour
arXiv:2602.02331v2 Announce Type: replace-cross Abstract: Achieving highly dynamic parkour with humanoid robots on complex terrains unseen during the pre-training stage remains a challenge in robotics. Although general l…
- arXiv cs.AIresearchFlow Policy Gradients for Robot Control
arXiv:2602.02481v2 Announce Type: replace-cross Abstract: Likelihood-based policy gradient methods are the dominant approach for training robot control policies from rewards. These methods rely on differentiable action l…
- arXiv cs.AIagentsDRMOT: A Dataset and Framework for RGBD Referring Multi-Object Tracking
arXiv:2602.04692v3 Announce Type: replace-cross Abstract: Referring Multi-Object Tracking (RMOT) aims to track specific targets based on language descriptions, playing a vital role in interactive AI systems such as robot…
- arXiv cs.AIresearchCoupled Local and Global World Models for Efficient First Order RL
arXiv:2602.06219v3 Announce Type: replace-cross Abstract: World models offer a promising avenue for more faithfully capturing complex dynamics, including contacts and non-rigidity, as well as complex sensory information,…
- arXiv cs.AIresearchOn the Dynamics of Delayed Generalization in Linear Models
arXiv:2602.08302v3 Announce Type: replace-cross Abstract: Delayed generalization, where training accuracy becomes nearly perfect substantially before test accuracy, has often been studied in the context of the depth and…
- arXiv cs.AIresearchCombining scEEG and PPG for reliable sleep staging
arXiv:2602.15042v2 Announce Type: replace-cross Abstract: Reliable sleep staging remains challenging for lightweight wearable devices such as single-channel electroencephalography (scEEG) or photoplethysmography (PPG). s…
- arXiv cs.AIresearchArticulated 3D Scene Graphs for Open-World Mobile Manipulation
arXiv:2602.16356v2 Announce Type: replace-cross Abstract: Semantic scene understanding has enabled robots to reason and perform affordance-driven object interaction. Despite this, real-world manipulation remains fundamen…
- arXiv cs.AIagentsFormal Policy Enforcement for Agentic Systems
arXiv:2602.16708v4 Announce Type: replace-cross Abstract: As agentic computing systems grow increasingly popular, they raise serious security concerns due to their susceptibility to various attacks. While prior research…
- arXiv cs.AIresearchPrompt Interference in LLM Post-Training: Why Pass@k Optimization Can Degrade Pass@1
arXiv:2602.21189v3 Announce Type: replace-cross Abstract: Pass@$k$ is a widely used performance metric for verifiable large language model tasks, including mathematical reasoning, code generation, and short-answer reason…
- arXiv cs.AIresearchAn Analysis of Role Semantics in Directed Dynamic Graph Modeling for Edge Classification
arXiv:2602.23135v3 Announce Type: replace-cross Abstract: Edge classification on directed dynamic graphs is critical for understanding evolving user-item interactions and behavioral patterns in recommender systems and so…
- arXiv cs.AIresearchToward Guarantees for Clinical Reasoning in Vision Language Models via Formal Verification
arXiv:2602.24111v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) show promise in drafting radiology reports, yet they frequently suffer from logical inconsistencies, generating diagnostic impressio…
- arXiv cs.AIresearchLearning Shortest Paths with Generative Flow Networks
arXiv:2603.01786v2 Announce Type: replace-cross Abstract: In this paper, we present a novel learning framework for finding shortest paths in graphs utilizing Generative Flow Networks (GFlowNets). First, we examine theore…
- arXiv cs.AIresearchCodecFlow: Efficient Bandwidth Extension via Conditional Flow Matching in Neural Codec Latent Space
arXiv:2603.02022v3 Announce Type: replace-cross Abstract: Speech Bandwidth Extension improves clarity and intelligibility by restoring/inferring appropriate high-frequency content for low-bandwidth speech. Existing metho…
- arXiv cs.AIresearchA Counterfactual Graph Approach to Oppositional Narrative Analysis
arXiv:2603.06135v3 Announce Type: replace-cross Abstract: Distinguishing legitimate critical speech from coordinated conspiracy narratives is a pressing challenge for the integrity of the online information ecosystem, ye…
- arXiv cs.AIresearchNeural-Surrogate Co-Design of Aerodynamic Shape and Control
arXiv:2603.06760v2 Announce Type: replace-cross Abstract: Control co-design jointly optimizes a robot's physical design and its controller. Existing techniques are limited in the number of model parameters and the comple…
- arXiv cs.AIresearchScheduling Parallel Optical Circuit Switches for AI Training
arXiv:2603.07373v2 Announce Type: replace-cross Abstract: The rapid growth of AI training has dramatically increased datacenter traffic demand and energy consumption, which has motivated renewed interest in optical circu…
- arXiv cs.AIresearchHybridStitch: Token-Adaptive Model Stitching for Efficient Diffusion Generation
arXiv:2603.07815v3 Announce Type: replace-cross Abstract: Iterative diffusion generation requires different amounts of model capacity at different token positions. Prior model combinations allocate capacity by denoising…
- arXiv cs.AIagentsGrammar of the Wave: Towards Explainable Multivariate Time Series Event Detection via Neuro-Symbolic VLM Agents
arXiv:2603.11479v5 Announce Type: replace-cross Abstract: Time Series Event Detection (TSED) aims to localize semantically meaningful events in time series data, with critical applications in high-stakes domains. Unlike…
- arXiv cs.AIagentsOxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism
arXiv:2603.14371v5 Announce Type: replace-cross Abstract: Embodied AI agents increasingly require parallel execution of multiple tasks, such as manipulation, conversation, and memory construction, from shared observation…
- arXiv cs.AIresearchAnolisVLA: Lightweight Vision-Language-Action Model with Deep State Space Models for Mobile Manipulation
arXiv:2603.15046v2 Announce Type: replace-cross Abstract: In this study, we address the problem of language-guided robotic manipulation, where a robot is required to manipulate a wide range of objects based on visual obs…
- arXiv cs.AIresearchDyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
arXiv:2603.16157v2 Announce Type: replace-cross Abstract: While Reinforcement Learning (RL) enhances Large Language Model reasoning, on-policy algorithms like GRPO are sample-inefficient as they discard past rollouts. Ex…
- arXiv cs.AIresearchContrastive Metric Learning for Point Cloud Segmentation in Highly Granular Detectors
arXiv:2603.23356v2 Announce Type: replace-cross Abstract: We propose a novel clustering approach for point-cloud segmentation based on supervised contrastive metric learning (CML). Rather than predicting cluster assignme…
- arXiv cs.AIagentsTopoPilot: Reliable Conversational Workflow Automation for Topological Data Analysis and Visualization
arXiv:2603.25063v2 Announce Type: replace-cross Abstract: Recent agentic systems demonstrate that large language models can generate complex codebases from natural language prompts. Despite this progress, they present re…
- arXiv cs.AIagentsJAL-Turn: Joint Acoustic-Linguistic Modeling for Real-Time and Robust Turn-Taking Detection in Full-Duplex Spoken Dialogue Systems
arXiv:2603.26515v2 Announce Type: replace-cross Abstract: Efficient and robust turn-taking detection remains a significant challenge in industrial-grade voice AI agents. Existing systems often rely on acoustic or linguis…
- arXiv cs.AIagentsIMAGAgent: Orchestrating Multi-Turn Image Editing via Constraint-Aware Planning and Reflection
arXiv:2603.29602v2 Announce Type: replace-cross Abstract: Existing multi-turn image editing paradigms are often confined to isolated single-step execution. Due to a lack of context-awareness and closed-loop feedback mech…
- arXiv cs.AIresearchSpec-Harness: Assessing and Guiding LLM-Synthesized Formal Specifications Beyond Verifiability
arXiv:2604.00280v3 Announce Type: replace-cross Abstract: Formal specifications play a central role in ensuring software reliability, yet automatically synthesizing high-quality specifications remains difficult and often…
- arXiv cs.AIresearchProcess Rewards for Outcome-Guided Reasoning Steps in LLM Reinforcement Learning
arXiv:2604.02341v2 Announce Type: replace-cross Abstract: Mathematical reasoning in large language models has improved substantially with reinforcement learning using verifiable rewards, where final answers can be checke…
- arXiv cs.AIresearchPickalo: Leveraging 6D Pose Estimation for Low-Cost Industrial Bin Picking
arXiv:2604.04690v3 Announce Type: replace-cross Abstract: Bin picking in real industrial environments remains challenging due to severe clutter, occlusions, and the high cost of traditional 3D sensing setups. We present…
- arXiv cs.AIresearchGenerating Synthetic Doctor-Patient Conversations for Long-form Audio Summarization
arXiv:2604.06138v2 Announce Type: replace-cross Abstract: Long-context audio reasoning is underserved in both training data and evaluation. Existing benchmarks target short-context tasks, and the open-ended generation ta…
- arXiv cs.AIresearchCMP: Robust Whole-Body Tracking for Loco-Manipulation via Competence Manifold Projection
arXiv:2604.07457v2 Announce Type: replace-cross Abstract: While decoupled control schemes for legged mobile manipulators have shown robustness, learning holistic whole-body control policies for tracking global end-effect…
- arXiv cs.AIresearchPR-MaGIC: Prompt Refinement Via Mask Decoder Gradient Flow For In-Context Segmentation
arXiv:2604.12113v2 Announce Type: replace-cross Abstract: Visual Foundation Models (VFMs) such as the Segment Anything Model (SAM) have significantly advanced broad use of image segmentation. However, SAM and its variant…
- arXiv cs.AIresearchFeed-Forward 3D Scene Modeling: A Problem-Driven Perspective
arXiv:2604.14025v2 Announce Type: replace-cross Abstract: Reconstructing 3D representations from 2D inputs is a fundamental task in computer vision and graphics, serving as a cornerstone for understanding and interacting…
- arXiv cs.AIagentsDive into Claude Code: The Design Space of Today's and Future AI Agent Systems
arXiv:2604.14228v3 Announce Type: replace-cross Abstract: Claude Code is an agentic coding tool that can run shell commands, edit files, and call external services on behalf of the user. This study describes its architec…
- arXiv cs.AIresearchBeyond Single-Model Optimization: Preserving Plasticity in Continual Reinforcement Learning
arXiv:2604.15414v3 Announce Type: replace-cross Abstract: Continual reinforcement learning must balance retention with adaptation, yet many methods still rely on \emph{single-model preservation}, committing to one evolvi…
- arXiv cs.AIagentsNegative Advantages Is a Double-Edged Sword: Calibrating advantages in GRPO for Search Agents
arXiv:2604.18235v3 Announce Type: replace-cross Abstract: Search agents achieve strong question-answering performance through multi-turn interactions with search engines, with Group Relative Policy Optimization (GRPO) be…
- arXiv cs.AIresearchTurboEvolve: Towards Fast and Robust LLM-Driven Program Evolution
arXiv:2604.18607v2 Announce Type: replace-cross Abstract: LLM-driven program evolution can discover high-quality programs, but high API cost and run-to-run variance hinder reliable progress. We propose TurboEvolve, a mul…
- arXiv cs.AIresearchUniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling
arXiv:2604.19734v2 Announce Type: replace-cross Abstract: Scaling humanoid foundation models is bottlenecked by the scarcity of robotic data. While massive egocentric human data offers a scalable alternative, bridging th…
- arXiv cs.AIresearchMILP the Gap: An Automated Reasoning Approach to Condorcet's Paradox
arXiv:2604.19851v2 Announce Type: replace-cross Abstract: Computational social choice studies how conflicting individual preferences are aggregated into a collective decision, with applications to elections, AI evaluatio…
- arXiv cs.AIresearchCall-Chain-Aware LLM-Based Test Generation for Java Projects
arXiv:2604.22046v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently shown strong potential for generating project-level unit tests. However, existing state-of-the-art approaches primarily…
- arXiv cs.AIresearchPermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training
arXiv:2604.22117v3 Announce Type: replace-cross Abstract: Aligned large language models (LLMs) remain vulnerable to adversarial manipulation, and their reliance on web-scale pretraining creates a subtle but consequential…
- arXiv cs.AIagentsCross-Stage Coherence in Hierarchical Driving VQA: A Consistency Protocol and Parameter-Efficient Gated Context Routing
arXiv:2604.22560v2 Announce Type: replace-cross Abstract: Graph Visual Question Answering (GVQA) for autonomous driving organizes reasoning into ordered stages, namely Perception, Prediction, and Planning, where planning…
- arXiv cs.AIresearchHiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models
arXiv:2604.23238v3 Announce Type: replace-cross Abstract: Distillation via reasoning traces exposes frontier models to adversarial third parties, who can bypass their guardrails and misappropriate their capabilities. Exi…
- arXiv cs.AIagentsStructure Liberates: How Constrained Sensemaking Produces More Novel Research Output
arXiv:2605.00557v2 Announce Type: replace-cross Abstract: Scientific discovery is an extended process of ideation -- surveying prior work, forming hypotheses, and refining reasoning -- yet existing LLM research agents tr…
- arXiv cs.AIresearchDependency Parsing Across the Resource Spectrum: Evaluating Architectures on High and Low-Resource Languages
arXiv:2605.02608v3 Announce Type: replace-cross Abstract: Transformer-based models achieve state-of-the-art dependency parsing for high-resource languages, yet their advantage over simpler architectures in low-resource s…
- arXiv cs.AIresearchRefining Compositional Diffusion for Reliable Long-Horizon Planning
arXiv:2605.03075v2 Announce Type: replace-cross Abstract: Compositional diffusion planning generates long-horizon trajectories by stitching together overlapping short-horizon segments through score composition. However,…