Hugging Face — daily papers
Written by: Fatih · Last updated: 2026-10-24
Last checked: 2026-10-24 · Monitoring since: 2026-10-01 · 404 records
What is monitored
From the api/daily_papers endpoint we take the identifier, title and date of the day's highlighted research papers (first 10).
What is not monitored
We do not take paper text or abstracts. Being listed does not mean peer-reviewed correctness; it only shows visibility.
Links
The API response has no address field — records carry no per-record link; papers are reachable via the root address. Source: https://huggingface.co/papers
Where this page's data comes from
The chain has four links and every address is written out here: the source's own page, the endpoint our engine actually calls, that endpoint's official documentation, and the source's terms of use. The endpoint address is compared against the address in the engine's code on every check; if the two diverge the gate turns red and publishing stops.
- Source page: https://huggingface.co/papers
- Endpoint called:
https://huggingface.co/api/daily_papers?limit=10 - Endpoint documentation: Hugging Face Hub API documentation
- Terms of use: Hugging Face terms of service
Observations
The first seen stamp on each row is the moment we first saw the record; it need not match the source's own publication date. The term is defined in the Glossary.
- — Different Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language Models (first seen 2026-10-24)
- — BrickWright: Agentic Creation of Physically Feasible Brick Builds (first seen 2026-10-24)
- — F2G-Pose: Geometry-Aware Foundation Feature Lifting for Direct RGB-D Category-Level Object Pose Estimation (first seen 2026-10-23)
- — AnyGameRenderer: Entity-Consistent Generative Rendering Across Games and an Entity-Level Benchmark (first seen 2026-10-23)
- — TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments (first seen 2026-10-24)
- — Agent Mosaic: Recognizing Confidentiality Leaks in Agentic LLM Systems (first seen 2026-10-24)
- — Self-Distilling Future Video Representations for World Action Models (first seen 2026-10-23)
- — Toward Recursive Self-Improvement in Physical AI (first seen 2026-10-23)
- — PACE: Periodic and Channel-Aligned Encoding of Four-Bit Recurrent States (first seen 2026-10-22)
- — RealMe: Immersive Volumetric Video from Minimal Captures (first seen 2026-10-22)
- — Think Harder Where It Matters: Item-Level Adaptive Recursion for Sequential Recommendation (first seen 2026-10-22)
- — Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use (first seen 2026-10-22)
- — DONIC: Generalist Dexterous Bimanual Tracking Control from Human Demonstrations (first seen 2026-10-22)
- — Diffusion Language Models Change Speed Across Tasks and Languages (first seen 2026-10-24)
- — HyperDrone: Simulation-to-Reality Transfer at Millions of Frames per Second (first seen 2026-10-24)
- — RLSplat: Resource-Linear Budget Control for Feed-forward 3D Gaussian Splatting (first seen 2026-10-23)
- — Automating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild? (first seen 2026-10-22)
- — Answering in the User's Variety: Calibrated Accommodation Pairs for Post-Training (first seen 2026-10-22)
- — Scaling in Games: Continued Pre-Training For Embodied Agents In Diverse Virtual Worlds (first seen 2026-10-22)
- — Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading (first seen 2026-10-21)
- — Multimodal Reranking Is a System One Decision (first seen 2026-10-21)
- — Harness Self-Evolution for Long-Horizon Terminal Agents (first seen 2026-10-21)
- — DecisionPivot: A Diagnostic Benchmark for Decision Models in LLM Systems (first seen 2026-10-21)
- — RouteWeaver: Weaving Mode Selection and Execution into Unified LLM Routing (first seen 2026-10-23)
- — Heartbeat Thinking: Speculative Reasoning and Tool Execution While Listening in Realtime Voice Agents (first seen 2026-10-23)
- — HarnessRL: Co-Evolving Policies with a Just-in-Time Training Harness for Open-Ended RL (first seen 2026-10-21)
- — More Memory, Worse Agents: Error Reproduction and Anti-Persistence in LLM Agents (first seen 2026-10-21)
- — SpecMRL: Redesign Matryoshka Representation Learning within Parametric Spectrum (first seen 2026-10-20)
- — Text-to-Image Generative Modeling Without Language Pretraining (first seen 2026-10-20)
- — HandsOn: A Large-Scale Simulation Benchmark for Generalizable Dexterous Manipulation (first seen 2026-10-20)
- — Vision-Language-Delta-Action: Learning Compact Deltas as Actionable Transition Interfaces for Robotic Manipulation (first seen 2026-10-20)
- — TILDE: Teacher-in-the-Loop Distillation from Embodied Foundation Models (first seen 2026-10-23)
- — Painting Change Map as guidance for generating painting processes (first seen 2026-10-21)
- — CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow? (first seen 2026-10-21)
- — Bridging Exploration and Persona: Information-Directed Dueling for Test-Time Agent Personalization (first seen 2026-10-24)
- — IF-MAP: Influence-Function-Inspired Agent Prioritization in Multi-Agent Systems (first seen 2026-10-24)
- — Editing Images into Reusable Layers via Compositional Edit-Prior Distillation (first seen 2026-10-23)
- — Good Judgment, Bad Advice: Decision-Model Controllers Raise Agent Cost Without Improving Success (first seen 2026-10-21)
- — HSPR: Recovering Software-Agent Benchmark Scores via Hierarchical Posterior Inference (first seen 2026-10-20)
- — EvalResearchBench: Can AI Agents Build Their Own Feedback for Recursive Self-Improvement? (first seen 2026-10-24)
- — The LLM-Attacker: A Unified Red-Team Framework for Privacy Evaluation of Split and Distributed LLM Systems (first seen 2026-10-24)
- — Entertain, then Cancel: How Thinking Makes Reasoning Models Read Some and Or Literally (first seen 2026-10-22)
- — Beyond Compilation: Evaluating Faithful Natural-Language-to-Lean Statement Formalization (first seen 2026-10-21)
- — LIBERO-Industry: Benchmarking Robotic Manipulation under Industrial Process Constraints (first seen 2026-10-20)
- — Mine, Sample, Render: Test-Time Diversity Alignment Of Generative Models (first seen 2026-10-17)
- — Behavioral Compositionality of Sparse Autoencoder Representations (first seen 2026-10-17)
- — LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents (first seen 2026-10-17)
- — PathFM: Fast Generative Forecasting of Multivariate Trajectories (first seen 2026-10-17)
- — CoDA: Co-Evolution of Data Engines and Action Policies (first seen 2026-10-16)
- — MJev-Spatial: System One Spatial Intelligence via Native Attention Matching (first seen 2026-10-16)
- — VLA-Δ: Learning Physical State Transition for Vision-Language-Action Models (first seen 2026-10-16)
- — Waypoint-OPD: Transferring RL-Induced Policy Shifts via Gated On-Policy Distillation (first seen 2026-10-16)
- — MMSearch-OPD: Easy Made Hard, Hard Made Learnable for Long-Horizon Multimodal Search Agents (first seen 2026-10-16)
- — MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering (first seen 2026-10-16)
- — τ-Emotion-Bench: Exposing the Interaction Fragility of LLM Agents through Real-World User Emotions (first seen 2026-10-15)
- — MolLean: A Process-Verifiable Language for Chemical Reasoning (first seen 2026-10-23)
- — PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis (first seen 2026-10-22)
- — BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References (first seen 2026-10-20)
- — Diminishing Returns of Data in Sparse Autoencoders: Functional Behavior does not Guarantee Feature Reproducibility (first seen 2026-10-17)
- — Equal CER, Unequal Damage: Severity-Weighted Character Error Rate for OCR (first seen 2026-10-17)
- — Learning to Read Out: Unembedding Dynamics in Language Model Pretraining (first seen 2026-10-17)
- — Latent Compaction with Test-Time Training (first seen 2026-10-15)
- — Cortex: Consolidating Context into Parametric Memory (first seen 2026-10-15)
- — VITAL: From Visual Motion to Precise Action (first seen 2026-10-15)
- — SafeDriveVLA: Navigation-Conditioned World Model Dreaming for Conflict-Aware End-to-End Autonomous Driving (first seen 2026-10-15)
- — Evidence-First Agentic Verification (first seen 2026-10-20)
- — LeanPRM: Lean-Grounded Process Supervision for Natural Language Proof Verification (first seen 2026-10-17)
- — Copy-Paste Language Models (first seen 2026-10-17)
- — Cowork-Gym: Synthesizing Interactive Work Environments at Scale (first seen 2026-10-16)
- — H-DDCM: Hadamard Codebooks for Fast Generative Image and Video Compression (first seen 2026-10-16)
- — SHELDON: Reasoning through Dual Search over Hypothesis and Observation Spaces (first seen 2026-10-14)
- — 4LC: 4D Latent Cache for Efficient Conditioning in Dynamic World Modeling (first seen 2026-10-14)
- — SWE-MISBEHAVIOR: Benchmarking and Mitigating Boundary Violations in Software Coding Agents (first seen 2026-10-14)
- — WorldRecon: Generative Multi-View Mesh Reconstruction for Large-Scale 3D World (first seen 2026-10-14)
- — Agent-Reflex: Teaching Language Agents to Act Reflectively (first seen 2026-10-14)
- — PINT: A Unified Model with Shared Predictive Intent for Robot Manipulation (first seen 2026-10-16)
- — Self-Distillation with Verbal Regularization (first seen 2026-10-14)
- — Scalable Weak-to-Strong Verifier Hardening for Agent Environments (first seen 2026-10-17)
- — SkillMarker: A KGW-Based Watermarking Method for Agent Skill Ownership Verification (first seen 2026-10-15)
- — Data-Centric Recursive Self-Improvement for Navigation via Policy-Synchronized Language-Action Dual Bootstrapping (first seen 2026-10-14)
- — EviScene: Beyond "Is It AI?" for Agentic Image Factuality Assessment (first seen 2026-10-14)
- — LightVAE: Towards Compact and Efficient Video Autoencoders (first seen 2026-10-14)
- — BigCell: Generating Gigapixel Whole-Slide Images (first seen 2026-10-15)
- — Latent World Action Model for GUI Agents (first seen 2026-10-14)
- — Learning More by Training Less: Uncertainty-Guided Progressive Contraction for Efficient Reasoning Post-Training (first seen 2026-10-13)
- — AgentSWE: Can Coding Agents Build the Agent You Actually Want? (first seen 2026-10-13)
- — Scaling Agentic Data for Long-Horizon Terminal Intelligence (first seen 2026-10-13)
- — PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling (first seen 2026-10-12)
- — Work2Skill: Capturing Human GUI Experience for Reusable Agent Skill Induction (first seen 2026-10-12)
- — Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding (first seen 2026-10-12)
- — The value of information and the emergence of communication in evolved e-puck robots (first seen 2026-10-12)
- — Continually Post-training Small Models via Synthetic Self-distillation (first seen 2026-10-12)
- — AI-Driven, Timing-Aware Power Optimization (first seen 2026-10-16)
- — Super-Resolution in The Right Latent Space: A Frozen Vision-Foundation Substrate (first seen 2026-10-15)
- — SWEeper-Bench: Can Agents Discover Bugs in Interactive Software? (first seen 2026-10-15)
- — Thinking Inertia: LLMs Keep Thinking When Told Not To (first seen 2026-10-13)
- — Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition (first seen 2026-10-12)
- — SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments (first seen 2026-10-12)
- — Reflex: Visually Grounded Reactive Humanoid Control (first seen 2026-10-12)
- — DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models (first seen 2026-10-12)
- — The Lattice of Transition Laws (first seen 2026-10-10)
- — Opera: A Verbal Critic Framework for Long-horizon Coding Agents (first seen 2026-10-10)
- — Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub (first seen 2026-10-10)
- — Predicting Cable Dynamics with Physical Attention Bias (first seen 2026-10-10)
- — MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination (first seen 2026-10-09)
- — MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement (first seen 2026-10-09)
- — Foundations of Large Language Models (first seen 2026-10-09)
- — OuroWorld: Bringing Any 3D World Alive as Diverse, Endlessly Looping 3D Cinemagraphs (first seen 2026-10-09)
- — BrickBench: Evaluating Agentic Brick Design (first seen 2026-10-09)
- — OmniCapBench: A Deep-Structured Evaluation Framework for Fine-Grained Audio-Visual Captioning (first seen 2026-10-09)
- — Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict (first seen 2026-10-09)
- — Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action Models (first seen 2026-10-20)
- — NavGPT-3: Harnessing Context in a Hierarchical Navigation Runtime (first seen 2026-10-15)
- — ARCS: Towards Precise Text-to-SQL via Structured Disambiguation (first seen 2026-10-13)
- — Evaluating the Transfer of Co-Evolved Communication from 2D to 3D Simulation (first seen 2026-10-10)
- — On-Policy Distillation Teaches New Skills but Not New Knowledge (first seen 2026-10-09)
- — From Pareto to Preference: Personalized Test-Time Scaling via Amortized Agentic Policy Discovery (first seen 2026-10-08)
- — QuadTok: Quadtree Visual Tokenizer for Autoregressive Image Generation (first seen 2026-10-08)
- — SkillForge: Co-Evolving Skills and Agents via Dynamic Skill Lifecycles (first seen 2026-10-08)
- — Tetris3D: 3D Scene Generation With Objects That Fit Together (first seen 2026-10-08)
- — SWE-Game: Can Coding Agents Build the Games We Want? (first seen 2026-10-08)
- — SPLATIFY: Reproduce, Discover, Innovate! From Papers and Ideas to Trainable 3DGS Code (first seen 2026-10-13)
- — A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning (first seen 2026-10-10)
- — Incidental information contaminates patient notes and disrupts clinical reasoning in large language models (first seen 2026-10-10)
- — Frozen Models, Evolving Expertise: Model-Agnostic Learning from Deployment Experience for Multimodal Medical AI (first seen 2026-10-10)
- — SPW-Nav: A Streaming Panoramic World Model for Language-Guided Navigation (first seen 2026-10-09)
- — PhysEvo: Astra Can Act, Let It (first seen 2026-10-08)
- — Co-Evolving Robot Orchestrators and Policies through Deployment (first seen 2026-10-08)
- — DAEDALUS: Bootstrapping Agent Memory from Self-Generated Tasks (first seen 2026-10-07)
- — CtrlCache: Accelerating Interactive Video World Models with Control-Aware Caching (first seen 2026-10-07)
- — MemPilot: Orchestrating On-Demand Multimodal Memory Curation for LLM Agents (first seen 2026-10-20)
- — Internalizing Agent Experience into Diffusion Model Weights via On-Policy Context Distillation (first seen 2026-10-08)
- — Conditional Trajectory Peaks: Single-Pass Multimodal Policies over Action Chunks (first seen 2026-10-07)
- — Execution-Aligned Progressive Noise for Consistent Asynchronous Replanning in Generative Robot Policies (first seen 2026-10-07)
- — WildMatch: Weakly Supervised Image Matcher Adaptation for Wildlife Re-Identification (first seen 2026-10-07)
- — Arm-wise Compositional Generalization in Dual-Arm Vision-Language-Action Models (first seen 2026-10-06)
- — OmniConfess: Eliciting Token Confessions to Mitigate Omni-Modal Hallucination (first seen 2026-10-06)
- — Noise Out, Bias In: Targeted Bias Injection in Diffusion Language Models via Closed-Loop Activation Steering (first seen 2026-10-06)
- — Certification of Real Images through Calibrated Content Authentication (first seen 2026-10-06)
- — Code2Games: Enabling Coding Agents for Gaming World Generation (first seen 2026-10-06)
- — Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancy (first seen 2026-10-06)
- — DiffGate: Difficulty-Gated Teacher Guidance for On-Policy Distillation (first seen 2026-10-07)
- — Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level Evaluation (first seen 2026-10-07)
- — Magic-W0: A Structured World-Action Foundation Model for Physical Intelligence (first seen 2026-10-07)
- — T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images (first seen 2026-10-13)
- — Gains and Collapse in On-Policy Distillation:A Reinforcement Learning Perspective (first seen 2026-10-08)
- — Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery (first seen 2026-10-06)
- — Dynamic Harness Search: Building Multi-Agent Systems Per-Query via Prediction (first seen 2026-10-06)
- — Native Action-Prior Learning from Videos for World Action Models (first seen 2026-10-05)
- — Pivot-SD: Efficient Self-Distillation for Masked Diffusion Language Models (first seen 2026-10-05)
- — HyperBrowseComp: A Multilingual and Multimodal Stress Test for Web-Browsing Agents (first seen 2026-10-05)
- — When Do Biological Reasoning Models Use Their Biological Inputs? (first seen 2026-10-12)
- — Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation (first seen 2026-10-08)
- — The Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models (first seen 2026-10-06)
- — From Retrieval to Typed Decisions: Calibrated System One Models from Biomedical Sentence Encoders (first seen 2026-10-05)
- — Local Support Learning (first seen 2026-10-05)
- — SimuVerity: Benchmarking Agents for Engineering-Grade Simulink Model Generation (first seen 2026-10-05)
- — Ego2Act: Evaluating Goal-Directed Manipulation in Egocentric Video Generation (first seen 2026-10-03)
- — Keyword Harnesses Fail Open: A Cheap Diagnostic Ladder for Tool-Use Claims in Small Language Models (first seen 2026-10-03)
- — InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation (first seen 2026-10-02)
- — SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation (first seen 2026-10-02)
- — GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed (first seen 2026-10-13)
- — GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives (first seen 2026-10-13)
- — Source Identification Is Not Fitness Testing: Measuring the Limits of Synthetic-Data Attribution (first seen 2026-10-07)
- — QuantCode Model: Specializing Language Models for Executable Algorithmic Trading Code (first seen 2026-10-06)
- — Does Learning Protein Folding Generalize to Broader Reasoning? (first seen 2026-10-05)
- — LexReward: A Taxonomy-Driven Reward Framework for Legal Language Models (first seen 2026-10-05)
- — Video Generation Models: A Survey of Post-Training and Alignment (first seen 2026-10-03)
- — Honeycomb: Constant-Size Scene Memory Representation for Video World Models (first seen 2026-10-03)
- — Memorizon: Training World Models Beyond Their Context Window (first seen 2026-10-02)
- — Benchmarking and Enhancing Skill-Level Memory for Partially Observable Robotic Manipulation (first seen 2026-10-02)
- — Smaller Models, Better Rejects: Preference Distillation Scaling (first seen 2026-10-02)
- — Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spaces (first seen 2026-10-02)
- — Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL (first seen 2026-10-02)
- — SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language Models (first seen 2026-10-01)
- — Soft Spatial Reasoning (first seen 2026-10-01)
- — Mitigating the Length-Scaling Tax with Online Distillation (first seen 2026-10-01)
- — How Much Is an AI Token Worth? Scaling Laws for Wild AI-Generated Web Text (first seen 2026-10-01)
- — LEAP: Learned Block-wise Evidence Retrieval for Long Audio-Video Perception (first seen 2026-10-01)
- — Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text (first seen 2026-10-01)
- — Behavioral Persistence and Incomplete Functional Transfer of Co-evolved Communication in Evolutionary Robotics (first seen 2026-10-10)
- — HelixWorld: A Real-time Interactive Audio-Visual World Model (first seen 2026-10-05)
- — MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation (first seen 2026-10-05)
- — MemFold: Learning Compact Soft Memory for Long-Context Personalization via On-Policy Optimization (first seen 2026-10-03)
- — Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It (first seen 2026-10-03)
- — Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling (first seen 2026-10-02)
- — Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs (first seen 2026-10-02)
- — TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion (first seen 2026-10-01)
- — Context Language Models (first seen 2026-09-30)
- — Chinese-Jev: Bringing System One Model to Chinese-Language Tasks (first seen 2026-09-30)
- — EpiCon: Collective Agent Learning through Co-Evolving Multimodal Memory (first seen 2026-09-30)
- — EngiWorld: What Can Frontier Agents Deliver in Professional Engineering Environments? (first seen 2026-09-30)
- — ReSPO: Reshaped Sequence Policy Optimization for Gradient Starvation in Off-Policy Learning (first seen 2026-10-09)
- — Persona Dosing: Calibrated Activation Steering for Graded Trait Control (first seen 2026-10-03)
- — Does Native 3D Texture Generation Necessarily Require 3D Assets for Training? (first seen 2026-10-02)
- — SeLMRoute: Probabilistic Semantic Evidence for Large Language Model Routing (first seen 2026-10-01)
- — Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual Evidence (first seen 2026-10-01)
- — NavHarness: Towards Lifelong Embodied Navigation (first seen 2026-10-01)
- — Same Bytes, Different Authority: Reserved-Token Representations in Chat-Template Prompt Injection (first seen 2026-09-30)
- — When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety (first seen 2026-09-29)
- — InfiniHand: Streaming World-Space Hand Motion Estimation from Egocentric Video (first seen 2026-09-29)
- — GeoVerse: World-Consistent Novel View Synthesis in Geometric Latent Space (first seen 2026-09-29)
- — Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning (first seen 2026-09-29)
- — On-Policy Self-Distillation for Multi-Turn Image Editing (first seen 2026-09-29)
- — Relevance Does Not Imply Applicability: Experience Activation for Personal GUI Agents (first seen 2026-10-13)
- — Pretraining Transformers with Quantized Softmax in Attention (first seen 2026-09-30)
- — When Privacy Moves ML-Mediated Decisions On Device: Information and Incentive Misalignment in Auctions (first seen 2026-09-29)
- — WideSWE: Can Coding Agents Coordinate Changes Across Repositories? (first seen 2026-09-29)
- — TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining (first seen 2026-09-29)
- — Replacing Large Language Models with Jev Decision Models for Low-Latency Edge Service Orchestration (first seen 2026-10-03)
- — X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization (first seen 2026-10-03)
- — OpenTumorBoard: A Real-World Benchmark of Multidisciplinary Tumor Board Discussion Trajectories (first seen 2026-10-03)
- — VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models (first seen 2026-09-30)
- — Residual Transferability in Neural Image Watermarking (first seen 2026-09-29)
- — Mara Chain: Rethinking Failure as a Stepping Stone for AI System Auto-Evolution (first seen 2026-10-10)
- — Omni-IO Skills: Harnessing Your Agent Omni-Native (first seen 2026-09-30)
- — Hyperspherical Semantic Trajectory Analysis: Mapping Technological Diffusion across Academic Preprints, Patent Signals, and Compute Scaling (first seen 2026-09-30)
- — ZooWork-ShopRanker: An Open, Preference-Aligned E-Commerce Reranker (first seen 2026-09-28)
- — Block Sparse Attention with Log-Linear Complexity (first seen 2026-09-28)
- — AgentWorld: Benchmarking Long-Horizon Collaboration of Multi-agent LLMs (first seen 2026-09-28)
- — InternW0-Δ: A World Action Model Bridging Predictive Dynamics and Actions with 20K+ Hours of Open Data (first seen 2026-09-28)
- — Accent Analogy Guidance: More Speaker Similarity at Equal Accent in Cross-Lingual Voice Cloning (first seen 2026-10-07)
- — Omni-Decision: Evidence-Ledger Planning for Omni-Modal Agents (first seen 2026-09-30)
- — How Reproducible Are Evaluation Conclusions? A Self-Audit of LLM-Inferred Prompt Structure (first seen 2026-09-29)
- — Jev in the Wild: A Data-Driven Analysis of the Jev Model's Functionality, Applications and Ecosystem (first seen 2026-09-28)
- — TrackEverything: Long Horizon Dense Tracking via De-Duplicating 3D Scene Representations (first seen 2026-09-28)
- — SLCA-GRPO: Resolving Cross-Segment Credit Misattribution in Tool-Calling RL (first seen 2026-09-28)
- — RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation (first seen 2026-09-26)
- — AV-GRPO: Modality-Anchored Decoupling Diffusion Reinforcement Learning for Joint Audio-Video Generation (first seen 2026-09-26)
- — Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs (first seen 2026-09-26)
- — Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures (first seen 2026-09-26)
- — Parts-of-Speech as Emergent Categories in SAE Latent Space (first seen 2026-09-26)
- — Coding Agents for Generalized Task and Motion Planning Problems (first seen 2026-09-26)
- — Rufus-Air: An Open LLM Post-Training Recipe (first seen 2026-09-25)
- — IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis (first seen 2026-09-25)
- — ViRDM: Taming Representation Distribution Matching for Few-Step Causal Video Generation (first seen 2026-09-25)
- — PUBG Ally: A Conversational Embodied Agent as an AI Teammate (first seen 2026-09-25)
- — World Action Agent: Harnessing VLMs for Robot Manipulation via World Action Rehearsal (first seen 2026-09-25)
- — WanPE: Towards Cinematic Prompt Enhancement for Modern Text-to-Video Generation (first seen 2026-09-25)
- — ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds (first seen 2026-09-25)
- — Learning to Discover Interesting Mathematics (first seen 2026-09-26)
- — DeltaWAM: Delta World Action Models for Bimanual Manipulation (first seen 2026-09-26)
- — On the Diffusibility of High-Dimensional Latents (first seen 2026-09-24)
- — All modalities are equal, but video is more equal: Closing the Cross-Attention Gap in Joint Video Generation (first seen 2026-09-24)
- — Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents (first seen 2026-09-24)
- — EmbodiedSWE: Coding Agents for Long Horizon Dexterous Robotics (first seen 2026-09-24)
- — StudentBench: AI and human tutoring yield equivalent GRE learning gains (first seen 2026-09-24)
- — MemBodied: Recurrent Associative Memory for Vision-Language-Action Models (first seen 2026-09-24)
- — FoMo: Forking Moment in Generative Trajectory as a Perceptual Distance (first seen 2026-09-28)
- — SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue (first seen 2026-09-24)
- — The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks (first seen 2026-09-23)
- — StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training (first seen 2026-09-23)
- — HappyWorld-Bench (first seen 2026-09-24)
- — Emergent Collusion in Long-Horizon LLM Agent Interaction (first seen 2026-09-23)
- — Lean Pool: An AI-Maintained Archive of Formalized Mathematics (first seen 2026-09-23)
- — All-in-One Multilingual Scene Text Recognition with Script-aware Mixture-of-Experts (first seen 2026-09-23)
- — Geometric and Semantic Coupling for Interaction Understanding in 3D Scenes (first seen 2026-09-23)
- — Circuit Hypernetworks for Quantum-Augmented Diffusion Language Models (first seen 2026-09-23)
- — GAE: Learning a Geometry-Native Latent Space for 3D-Consistent World Generation (first seen 2026-09-23)
- — 1% of Tokens Can Be Enough: On Gradient Estimation in On-Policy Distillation (first seen 2026-09-22)
- — onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction (first seen 2026-09-22)
- — Harness-Zero: Harness Distillation via Agent-as-Harness (first seen 2026-09-22)
- — GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay (first seen 2026-09-22)
- — Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents (first seen 2026-09-22)
- — VideoGen-Agent: Reinforcing Video Generation Agents (first seen 2026-09-22)
- — Think Like a World Model, Act Like a VLA: Distilling World-Model Representations into Compact Robot Policies (first seen 2026-09-22)
- — CARD: Cluster-level Adaptation with Reward-guided Decoding for Personalized Text Generation (first seen 2026-09-28)
- — Do Implicit Personalization and Explicit Styles Conflict? PsPLUG: A Lightweight Plug-in for Balancing Personalization and Style in Customized LLMs (first seen 2026-09-28)
- — OmniEcho: Spatial Audio Understanding for Embodied Agents (first seen 2026-09-25)
- — PackLab: A Comprehensive Framework for Developing, Training, and Evaluating MLLMs in Robotic Bin Packing (first seen 2026-09-24)
- — Grounded Action Model: 3D Grounding as a Foundation for Robotics (first seen 2026-09-22)
- — Neural Spectral Capacity: Measuring and Designing Architectures from Network Specification Alone (first seen 2026-09-25)
- — Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World (first seen 2026-09-24)
- — OmniEdu: Open Foundation Models for Learning and Teaching (first seen 2026-09-22)
- — Calibrating Teacher--Student Discrepancy for On-Policy Distillation (first seen 2026-09-21)
- — CodeMidas: Scaling Agentic Coding RL Environments from Code Itself (first seen 2026-09-21)
- — TeleAntiFraud 2.0: A Refreshable, Profile-Grounded, and Audio-Based Benchmark for Telecom Fraud Detection (first seen 2026-09-21)
- — Training-Adaptive Convolutional Sparse Coding via Information Bottleneck for Robust Visual Representation (first seen 2026-09-21)
- — Paint-Anything: Unified Any-Color Control for Image Generation and Editing (first seen 2026-09-21)
- — DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation (first seen 2026-09-21)
- — Self-Evolving Search Index (first seen 2026-09-19)
- — VākQA: A Benchmark and Evaluation Study for Telugu Spoken Factoid Question Answering (first seen 2026-09-19)
- — Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL (first seen 2026-09-19)
- — FAMOS: Feed-Forward 3D Articulation Modeling from Sparse Observations (first seen 2026-09-19)
- — When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation (first seen 2026-09-19)
- — ALPINE: Adaptive Localization for Parameter- and Sample-Efficient Few-Shot Learning (first seen 2026-09-23)
- — FRAUDSkill: Structured Frozen-Weight Skill Optimization for Audio Anti-Fraud Detection (first seen 2026-09-21)
- — Sample Count Is Not Enough: Candidate-Generation Strategy Shapes the Energy and Performance of LLM Test-Time Scaling (first seen 2026-09-19)
- — Can MiniMax-H3 Reason About the Physical World? An Evaluation of Omni-Modal Generative Model (first seen 2026-09-19)
- — PACT: Can Enterprise AI Assistants Be Trusted Under Pressure? (first seen 2026-09-19)
- — Verifiable Social Reasoning for LLM Assistants (first seen 2026-09-19)
- — Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems (first seen 2026-09-16)
- — Modality-Autoregressive World-Action Models (first seen 2026-09-16)
- — PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Control (first seen 2026-09-16)
- — AI for Games in the Foundation Model Era (first seen 2026-09-16)
- — Bellman Policy Optimization (first seen 2026-09-23)
- — MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup (first seen 2026-09-21)
- — EvoOntology: A Self-Evolving Ontology Layer for Data Agents (first seen 2026-09-21)
- — Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration (first seen 2026-09-16)
- — Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States (first seen 2026-09-16)
- — Disentangling Representation Evolution in Transformers through Directional Decomposition (first seen 2026-09-16)
- — RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments (first seen 2026-09-15)
- — HazardAuditor: From Executable Threats to Safer Computer-Use Agents (first seen 2026-09-15)
- — Pick Your Poison: Learning to Select Poison Sets for Stronger LLM Backdoor Attacks (first seen 2026-09-15)
- — Enabling Creative Exploration for Vibe Design Agents (first seen 2026-09-15)
- — When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis (first seen 2026-09-15)
- — Atria Dawn: The Dawn of Agentic Superintelligence (first seen 2026-09-15)
- — Kaininja: Extending Native 3D Generators to the Part Level (first seen 2026-09-15)
- — Drift-Constrained Optimization: Only Direction Matters in Fine-Tuning Instruct Models (first seen 2026-09-16)
- — Training Specialist Models without Reasoning Trajectories for Domain Expert Distillation (first seen 2026-09-16)
- — Realtime-Venus: A full-duplex interaction system with asynchronous delegation (first seen 2026-09-15)
- — Agent as Policy for Robotic Manipulation (first seen 2026-09-15)
- — ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search (first seen 2026-09-15)
- — Breaking the Vision-Action Shortcut: Latent Interface Training for Generalizable Robotics Foundation Models (first seen 2026-09-14)
- — SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image (first seen 2026-09-14)
- — SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking (first seen 2026-09-14)
- — The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement (first seen 2026-09-16)
- — Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation (first seen 2026-09-14)
- — COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization (first seen 2026-09-14)
- — Beyond Solver Verdicts: Generative Reward Models for Autoformalization (first seen 2026-09-12)
- — Memory as Plans: World-Action Modeling with Memory-Grounded Planning (first seen 2026-09-12)
- — Negative Self-Distillation: Learning to Reason by Avoiding Flaws (first seen 2026-09-12)
- — FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimation (first seen 2026-09-12)
- — X-AuT: Progressive Audio-Encoder Compression for Speech LLMs with Cross-Scale Distillation (first seen 2026-09-11)
- — Generative Late-Interaction Embeddings For Visual Document Retrieval (first seen 2026-09-11)
- — Recursive Code World Models: Building Complex Worlds through Recursive Scene Programs (first seen 2026-09-11)
- — World in World: Explore the World with World Models (first seen 2026-09-11)
- — SenseNova-U1.5: Towards Native Unified Visual Intelligence (first seen 2026-09-11)
- — Think Before You Link: Rarity, Reasoning, and Retrieval in Multilingual Entity Linking (first seen 2026-09-12)
- — Building Multilingual Bridges: Data Mixing as the Pillar of Generalization for In-Language Reasoning (first seen 2026-09-12)
- — IdeaAMBIG: Benchmarking Implementation-Critical Gaps in Research-Idea Specifications (first seen 2026-09-12)
- — An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics (first seen 2026-09-11)
- — NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction (first seen 2026-09-11)
- — RESCUE-BENCH: Towards Relation-Aware Multi-Party Emotional Support Conversation Systems (first seen 2026-09-10)
- — Studying Image Tokenizers as Visual Languages in Unified Multimodal Models (first seen 2026-09-12)
- — ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation (first seen 2026-09-12)
- — SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research? (first seen 2026-09-10)
- — Co-Evolving Harnesses and Models: On-Policy Correction Helps Weaker Models Catch Up Where Imitation Fails (first seen 2026-09-10)
- — Marigold V2: Revisiting Diffusion Transformers for Monocular Depth Estimation (first seen 2026-09-09)
- — SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem (first seen 2026-09-11)
- — DF26: We Cannot Tell Fake From Real Anymore (first seen 2026-09-10)
- — Scores Alone Do Not Prove Discovery: The Discovery Certification Protocol for Auditing AI Research Agents (first seen 2026-09-10)
- — RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting (first seen 2026-09-09)
- — Harnessing CLIP and DINO: An Uncertainty-Aware Cascaded Fusion Network for Generalizable Deepfake Image Detection (first seen 2026-09-09)
- — Encoded Early, Used Late: Where Transformers Begin to Act on an Inferred Partner's Expertise (first seen 2026-09-09)
- — Measuring Language Transfer in Robot Policies: Adding Greek to a Cosmos3 Vision-Language-Action Policy (first seen 2026-09-09)
- — SQS: Bayesian DNN Compression through Sparse Quantized Sub-distributions (first seen 2026-09-09)
- — ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding (first seen 2026-09-09)
- — Adaptive Bridge: A Proxy-Based Decoupling Layer for Mitigating DDS Backpressure in ROS 2 (first seen 2026-09-12)
- — DianShi-RxnDB: A Large-Scale, Fine-Grained Organic Reaction Data Platform Built via a Fully Automated Pipeline for Researchers and AI Agents (first seen 2026-09-10)
- — MLLMs Hallucinate when Information Distribution Drifts in Synergy Heads (first seen 2026-09-21)
- — EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents (first seen 2026-09-11)
- — Diffs vs. Whole Files: An Empirical Comparison of Iterative Edit-Based and Direct Generation for Flutter/Dart Code Models (first seen 2026-09-10)
- — Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model (first seen 2026-09-09)
- — Srijika: OpenType-Layout-Reusing Font Restyling for Nine Indic Scripts (first seen 2026-09-19)
- — WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data (first seen 2026-09-10)
- — RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks? (first seen 2026-09-09)
- — Ask Before You Optimize: Dynamic Pre-Formulation Clarification for Interactive Optimization (first seen 2026-09-07)
- — RISE: Recursive Improvement via Self-Extrapolating Policy Distillation (first seen 2026-09-07)
- — Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference (first seen 2026-09-07)
- — WorldSculpt: Generating Compositional Worlds from Grounded Videos (first seen 2026-09-07)
- — Unlocking Lossless Speedups in LLMs via Discrete Diffusion (first seen 2026-09-08)
- — FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience (first seen 2026-09-08)
- — What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation (first seen 2026-09-08)
- — Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning (first seen 2026-09-08)
- — HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals (first seen 2026-09-08)
- — AdaptVPR: Route-Aware Hard Positive Generation for Robust Visual Place Recognition (first seen 2026-09-07)
- — When Models Edit Too Much: On the Fidelity of Minimal Code Edits (first seen 2026-09-07)
- — MaxKernel: Agentic Kernel Generation for TPUs (first seen 2026-09-07)
- — When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference (first seen 2026-09-07)
- — DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training (first seen 2026-09-05)
- — Last Translation Benchmark (first seen 2026-09-05)
- — Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs (first seen 2026-09-05)
- — Scal3R: Learning Efficient Multi-Relative Pose Query for Scalable Online 3D Reconstruction (first seen 2026-09-04)
- — Compile by Training: Turning Natural-Language Specifications into Local Neural Functions (first seen 2026-09-04)
- — PACE: Towards Surfacing Hidden Conflicts in User Requests (first seen 2026-09-04)
- — Puffin-World: Scaling a Unified Multimodal Model with Native 3D World States (first seen 2026-09-04)
- — Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding (first seen 2026-09-04)
- — FlashRender: Few-Step Generative Rendering via Camera-Controlled Video MeanFlow (first seen 2026-09-04)
- — WorldReward: Reward Modeling for Camera-Conditioned World Models (first seen 2026-09-04)
- — Measuring the Checker: Mutation Analysis for GPU-Kernel Benchmark Oracles (first seen 2026-09-22)
- — Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation (first seen 2026-09-08)
- — ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding (first seen 2026-09-07)
- — RoboTok: An Internet-Scale Data Engine for Human Demonstration Retrieval and Dexterous Manipulation Learning (first seen 2026-09-05)
- — A Common Measure of Communication for Speech Brain-Computer Interfaces (first seen 2026-09-05)
- — VeriPhy: Agentic Physical Reasoning for World Model Evaluation and Refinement (first seen 2026-09-05)
- — Let Confidence Change, Not the Prediction: Prediction-Preserving Repair for Post-hoc Calibration (first seen 2026-09-04)
- — Percolation Dynamics in Optimization : Variance Cascades and Discrete Scale Invariance (first seen 2026-09-04)
- — The Missing Temporal Link: Temporal Context Routing for Script-Driven Audio-Video Generation (first seen 2026-09-04)
- — EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents (first seen 2026-09-08)
- — PLC-DPO: Posterior Label Correction in Noisy and Ambiguous Preference Optimization (first seen 2026-09-14)
- — Puppeteer: Object-Grounded Posture-Aware Co-Speech Gesture Generation (first seen 2026-09-10)
- — AgenticGen: Reward-Guided Agentic Video Generation for Advertising (first seen 2026-09-10)
- — Dr. Claw: An AI Scientist Workspace for Vibe Research (first seen 2026-09-08)
- — Using Grounded Theory for Agent Behavior Analysis at Scale (first seen 2026-09-05)
- — AgentKernel: The Trust-Native Agentic Operating System (first seen 2026-09-25)
- — Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questions (first seen 2026-09-09)
- — Locked at the Entrance, Open Inside: Where RLVR Narrows the Solution Space (first seen 2026-09-05)
- — QCell: Recombining and Aligning Cell Queries for Overlapping Instance Segmentation (first seen 2026-09-05)
- — HyQuant: Hybrid-Precision Quantization for LLM Attention (first seen 2026-09-11)
- — Motion-Omni: End-to-End Joint Speech and Full-Body Motion for Spoken Dialogue (first seen 2026-09-07)
- — Knowing When Not to Reuse: Conditional Experience Transfer in Autonomous LLM Post-Training (first seen 2026-09-04)
- — One Symptom, Three Levers: A Critical Review of On-Policy Self-Distillation (first seen 2026-09-08)
- — Real-World Knowledge-Guided Change Data Synthesis for Remote Sensing (first seen 2026-09-08)