AI 论文速览

最近采集到 61 篇论文，最新在前。

• cs.AI Coding Agents with an Obstacle-Aware Harness for Safe Robot Manipulation
  2026-09-17 · 摘要（英文）
• cs.AI Workspace Models: Lightweight Robotic Memory via Saliency-Driven Supervision
  2026-09-17 · 摘要（英文）
• cs.AI FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observations
  2026-09-17 · 摘要（英文）
• cs.AI Paint-Anything: Unified Any-Color Control for Image Generation and Editing
  2026-09-17 · 摘要（英文）
• cs.AI ERCPMP-Gx: Endoscopic Image and Video Dataset for Morphological, Histopathological, and Genomic Characterization of Colorectal Polyposis
  2026-09-17 · 摘要（英文）
• cs.AI Quantifying Overclaiming Propensity in Frontier LLM Agents
  2026-09-17 · 摘要（英文）
• cs.AI An Empirical Study of Harness Design for Coding Agents
  2026-09-17 · 摘要（英文）
• cs.AI RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning
  2026-09-17 · 摘要（英文）
• cs.AI Harm Laundering in GPT Models: Evidence That Gender Discrimination Is Transformed Rather Than Reduced Across Safety-Trained Generations
  2026-09-17 · 摘要（英文）
• cs.AI GeoAAC: Geometry-Based Adaptive Action Chunking from Denoising Trajectories in VLA Policies
  2026-09-17 · 摘要（英文）
• cs.AI Semantic Action Graph: A Shared Representation for Agent Grounding and Human Interpretation of Sports Highlights
  2026-09-17 · 摘要（英文）
• cs.AI Prediction-Powered Smoothing and Validation for Disaggregated AI Evaluation
  2026-09-17 · 摘要（英文）
• cs.AI RAFT: A Stateful Retrieval-Augmented Framework for Troubleshooting Agents
  2026-09-17 · 摘要（英文）
• cs.AI Large Language Models as Falsifiers for Cyber-Physical Systems
  2026-09-17 · 摘要（英文）
• cs.AI Q&A on Any Spreadsheet Requires Interpreting Its Grid Structure
  2026-09-17 · 摘要（英文）
• cs.AI Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models
  2026-09-17 · 摘要（英文）
• cs.AI Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL
  2026-09-17 · 摘要（英文）
• cs.AI HIL-UMI: Bringing Human-in-the-Loop Post-Training of Vision-Language-Action Models to Universal Manipulation Interface
  2026-09-17 · 摘要（英文）
• cs.AI Ownership in AI-Assisted Everyday Tasks
  2026-09-17 · 摘要（英文）
• cs.AI PAA: The Probabilistic Allen Algebra: A Generative and Complete Probabilistic Extension of Allen's Interval Relations
  2026-09-17 · 摘要（英文）
• cs.LG Embedding Models Measure in Peculiar Ways
  2026-09-17 · 摘要（英文）
• cs.LG How Does Distribution Shift Shape Pretraining Gains in Neural PDE Surrogates?
  2026-09-17 · 摘要（英文）
• cs.LG Score Centering Stabilizes Off-policy Reinforcement Learning
  2026-09-17 · 摘要（英文）
• cs.LG PosteriorBench: From Point Estimates to Posterior Matching in Evaluating Generative Inverse Solvers
  2026-09-17 · 摘要（英文）
• cs.LG Calibrated RF-Fingerprinting Under Interference With Heterogeneous Transmission Protocols
  2026-09-17 · 摘要（英文）
• cs.LG Agile-WAM: An Agile Tactile World Action Model for Contact-Rich Robot Control
  2026-09-17 · 摘要（英文）
• cs.LG OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher
  2026-09-17 · 摘要（英文）
• cs.LG dQwen3.5: Hybrid-Attention Diffusion Language Models
  2026-09-17 · 摘要（英文）
• cs.LG MILER: Semantic Mid-Level Representation for Sim-to-Real Reinforcement Learning in Unstructured Autonomous Driving
  2026-09-17 · 摘要（英文）
• cs.LG Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation
  2026-09-17 · 摘要（英文）
• cs.LG Stable Movement for Nondual Lipschitz Convex Optimization: Efficiency and Nearly Optimal Oracle Rates
  2026-09-17 · 摘要（英文）
• cs.LG TetrisCNN for interpretable detection of phases of matter from experimental quantum simulator data
  2026-09-17 · 摘要（英文）
• cs.LG The First-Order Oracle Complexity of Lipschitz Convex Optimization in Nondual Settings
  2026-09-17 · 摘要（英文）
• cs.LG RISC-V and machine learning: a survey
  2026-09-17 · 摘要（英文）
• cs.CL Unifying Models of Intergroup Hostility in Online Discourse
  2026-09-17 · 摘要（英文）
• cs.CL JEPA-Anything: Learning Predictive Models across Different Worlds
  2026-09-17 · 摘要（英文）
• cs.CL On-Demand Attention: Language Models Know When to Recall
  2026-09-17 · 摘要（英文）
• cs.CL Summarization Bias: The Directional Collapse of Objective Projection into Told-Mode Labels in Large Language Models --- A Conceptual Framework and Registered Test Protocol
  2026-09-17 · 摘要（英文）
• cs.CL HerHealthEval: Evaluating Multilingual and Register-Sensitive Understanding of Women's Health Communication
  2026-09-17 · 摘要（英文）
• cs.CL UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising
  2026-09-17 · 摘要（英文）
• cs.CL Chronicle: Cut-Point Replay for Regression Testing of LLM Agents
  2026-09-17 · 摘要（英文）
• cs.CL What Does Privileged Information Add to On-Policy Self-Distillation?
  2026-09-17 · 摘要（英文）
• cs.CL WiC is Not WSD: A Study on LLMs and Lexical Ambiguity Resolution
  2026-09-17 · 摘要（英文）
• cs.CL SAFARI: An Industrial Benchmark for LLM-Assisted Hazard Analysis and Risk Assessment
  2026-09-17 · 摘要（英文）
• cs.CL Steering the Compass: Aligning Dynamic Psychological Counseling Conversations with Cognitive Behavioral Therapy Strategies
  2026-09-17 · 摘要（英文）
• cs.CL Language-model groups overstate consensus when replaying human deliberation on a reasoning task
  2026-09-17 · 摘要（英文）
• cs.CV Can 4D Foundation Models Remember?
  2026-09-17 · 摘要（英文）
• cs.CV SplashSplat: Reconstructing Splashing Liquids from Real-World Multi-View Videos
  2026-09-17 · 摘要（英文）
• cs.CV FlowSGS: Improving Flow Matching Priors for Inverse Imaging with Stochastic Interpolants
  2026-09-17 · 摘要（英文）
• cs.CV Should This Case Be Adapted? Prediction Fragmentation Controls Test-Time Adaptation
  2026-09-17 · 摘要（英文）
• cs.CV Towards Scaling Marine Perception with Synthetic Data
  2026-09-17 · 摘要（英文）
• cs.CV FunArt: Decoding Functional Structure and Articulation from Generative 3D Latents
  2026-09-17 · 摘要（英文）
• cs.CV Learning Foresight without Explicit Trajectories for 3D Diffusion Policies
  2026-09-17 · 摘要（英文）
• cs.CV Earth Surface Immune System for Rapid Monitoring of Unknown Anomalies
  2026-09-17 · 摘要（英文）
• cs.CV DexTouch-WM: Learning Action-Conditioned Tactile World Models from Human Touch for Dexterous Robot Manipulation
  2026-09-17 · 摘要（英文）
• cs.CV PROVIA: Procedure State Tracking for Online Mistake Detection in Egocentric Videos
  2026-09-17 · 摘要（英文）
• cs.CV Refinement Is Inherently Editable: Training-Free Prompt-to-Prompt Image Editing with Generative Refinement Network
  2026-09-17 · 摘要（英文）
• cs.CV PhGS: Post-Hoc Pruning and Refinement of Single-View Feed-Forward 3D Gaussian Reconstructions
  2026-09-17 · 摘要（英文）
• cs.CV INSPECT: Learning Robot View Selection from Assistant Use
  2026-09-17 · 摘要（英文）
• cs.CV RawSLAM: Online HDR Gaussian SLAM from Linear Radiance
  2026-09-17 · 摘要（英文）
• cs.CV CoRef-GS: Cooperative Referring Gaussian Splatting for Multi-Agent Scene Understanding
  2026-09-17 · 摘要（英文）