All papers
Every paper that carries a page of its own here. The Papers page leads with what is trending and lets you filter the recent catalog; this is the plain index of the rest.
Reflect, Revise, Reuse: Training-Free Skill Evolution for GUI Agents2026Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents2026The Other Half of the Memory Wall: Serving 35B MoEs from SSD with Trained Routing Prediction2026Agora: Git as Shared Memory for Collective AutoResearch2026ActionPiece: Rethinking Action Tokenization for Autoregressive Vision-Language-Action Models2026PACT: Can Enterprise AI Assistants Be Trusted Under Pressure?2026Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening2026CERA-MoA: Co-Evolving Routing Mechanisms with Continually Learning LLM Agents2026Fathom: Per-Query Read Depth for Sparse Decoding over Offloaded KV Caches2026TabPFN-3.5: Technical Report2026Zing-0.5: Toward Playable Worlds with Real-Time Joint Action and Text Control2026Gaze as Evidence for Common Grounding: A Cross-Corpus Analysis of MapTask and MUNDEX2026ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments2026PANORAMA: Panoptic Grounded Captioning via Mask Proposal Selection2026DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression2026What Does Privileged Information Add to On-Policy Self-Distillation?2026RetireOPD: Self-Retiring On-Policy Distillation for Agentic Reinforcement Learning2026JEPA-Anything: Learning Predictive Models across Different Worlds2026Prediction-Powered Smoothing and Validation for Disaggregated AI Evaluation2026Self-Evolving Search Index2026When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation2026Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation2026SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness2026BI-Agent and BI-Bench: Towards Automating End-to-End Business Intelligence2026Geometry of Values: Task Vector Composition for Ethical Preference Alignment in Language Models2026OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual Dialogue2026Training-Adaptive Convolutional Sparse Coding via Information Bottleneck for Robust Visual Representation2026When AI Reviews Train AI Reviewers: Scientific-Judgment Collapse and Mitigation2026IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts2026GraphSkillEvo: Evolutionary Optimization of Graph-Structured Agent Skills2026RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents2026TCDA: Thread-Constrained Discourse-Aware Modeling for Conversational Sentiment Quadruple Analysis2026AffordanceWAM: Affordance-Aware Joint World-Action Modeling for Robot Manipulation2026RewardVerse: Rubric-Guided Policy Optimization for Video Reward Modeling2026Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World2026Neural Spectral Capacity: Measuring and Designing Architectures from Network Specification Alone2026OmniEcho: Audio-Visual Spatial Understanding for Omni-Modal Embodied Agents2026Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents2026ACLArena: Agent Continue Learning in Multi-stage Post-training2026$t_0$: A Time-Series Foundation Model for Forecasting with Context2026Uranus: Building the Next-Generation Simulation Infrastructure for Embodied AI2026Emergent Collusion in Long-Horizon LLM Agent Interaction2026RRSI: Regularized Recursive Self-Improvement of Agent Harnesses2026Harness-Zero: Harness Distillation via Agent-as-Harness2026WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory2026GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay2026WaveFront Decoding: Parallelized Self-Speculative Decoding for Looped Language Models2026Why Do Video Diffusion Models Violate Physics? Unveiling the Flaws in Attention Mechanisms20261% of Tokens Can Be Enough: On Gradient Estimation in On-Policy Distillation2026OmniEdu: Open Foundation Models for Learning and Teaching2026onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction2026LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay2026Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings2026From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health2026X-Planner: Event-Structured Task Planning for Embodied Intelligence2026Lean Pool: An AI-Maintained Archive of Formalized Mathematics2026Geometric and Semantic Coupling for Interaction Understanding in 3D Scenes2026The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks2026D-JEPA: A Decision-Aligned Latent World Model2026Complex KDA: Understanding and Enhancing the Expressivity of Kimi Delta Attention2026Disaggregated Quantization: Specializing LLM Prefill and Decode2026MemoryAthena: Adaptive Routing over Latent and Generated Memories2026Blaming Across the Aisle: Political Contrasting and Blame Attribution in the Danish Parliament2026Knowledge Pull Requests for Continual Document Authoring2026SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue2026Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs2026WhatWorkedBench: Benchmarking Experimental Understanding in AI Agents2026MemBodied: Recurrent Associative Memory for Vision-Language-Action Models2026Agent-Editing World Model: Rethinking World Modeling for LLM Agents2026StudentBench: AI and human tutoring yield equivalent GRE learning gains2026GTR: Gated Token Recurrence for Efficient Dense Prediction2026Six Layers Less: Encoder Pruning for Whisper with Label-Free Recovery2026LastOPD: Taming Collapse in Latent On-Policy Distillation2026Post-Training Leaves Behavioral Shadows on Unrelated Decisions2026Parts-of-Speech as Emergent Categories in SAE Latent Space2026Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures2026IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis2026Training Object Permanence in World Models2026Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents2026SAGE: Mitigating Long-Horizon Reasoning Biases via Topological Guidance2026TrackEverything: Long Horizon Dense Tracking via De-Duplicating 3D Scene Representations2026Coding Agents for Generalized Task and Motion Planning Problems2026Rolling-WAM: World Action Models with Rolling Imagination2026G$^2$PTQ: Improving LLM Post-Training Quantization with Generalized Gradient Compensation2026Game Arena: Strategic LLM Evaluation in Competitive Environments2026TRACE: Temporal Audit and Condition-aware Evaluation of Streaming Video Understanding2026CaptchaArena: A Large-Scale, Fine-Grained Dataset for Training Computer-Use Agents on Interactive CAPTCHAs2026Receiver-Conditioned Latent Communication gives 94% CacheBack2026PrismQuant: Optimal Null-Space Rotations for Grouped Quantizers2026Rethinking Training-Inference Mismatch in LLM Reinforcement Learning: Where It Arises and How to Correct It2026DepthBench: Measuring How Residual Connections Enable More Computational Depth2026PrefixGuard: Online Failure Warning and Trace-Grounded Diagnosis for LLM Agents2026Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction2026Omni-IO Skills: Harnessing Your Agent Omni-Native2026AdaTutoRank: Learning to Rerank Document Sets via Adaptive Tutoring Optimization for RAG and Deep Research2026ExpVoyager: Direct Experience Navigation for Dynamic Agent Skill Synthesis2026MassAlloc Attention: Let Attention Allocate Its Own Compute2026OpenTumorBoard: A Real-World Benchmark of Multidisciplinary Tumor Board Discussion Trajectories2026TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces2026Raven: The Harness of Harnesses for Composable Agentic Intelligence2026Program-Verified Self-Evolution for Vision-Language Models2026Opera: A Verbal Critic Framework for Long-horizon Coding Agents2026BIABench: Evaluating AI agents on real-world bioimage analysis tasks2026Knowing When Thinking Is Not Enough: Teaching Small Reasoning Models to Reason Beyond Their Parametric Knowledge2026Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual Evidence2026Draft-KV: Learning Useful Latent Communication Between Language Models2026ColNanoVDR: Document-Free Query Distillation for Multi-Vector Visual Document Retrieval via Optimal Transport2026Understanding On-Policy Distillation: A Mechanistic Interpretability Perspective via Sparse Crosscoders2026LLMs are General Asynchronous Agents2026Rubric Rewards from Item Response Theory2026Improving Test-Time Scaling with Adaptive Looped Transformers2026CUA-SWE: When Computer-Use Agents Meet Visual Software Engineering2026RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied Agents2026AgentTell: Behavioural Side-Channel Leakage in Browser-Use Agents2026X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization2026VGGT-Diff: Visual Geometry Meets Diffusion for Sparse-View Novel View Synthesis2026MemFold: Learning Compact Soft Memory for Long-Context Personalization via On-Policy Optimization2026Triadic Linear Attention: Three-Dimensional Recurrent States for Long-Context Sequence Modeling2026Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It2026Can Agents Design Libraries for Agents?2026CorrGRPO: Correlation-Normalized GRPO for Multi-Reward Learning2026Chinese-Jev: Bringing System One Model to Chinese-Language Tasks2026Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents2026EngiWorld: What Can Frontier Agents Deliver in Professional Engineering Environments?2026Context Language Models2026Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI2026Beyond the Timeline: Augmenting Long-Video Memory with Grounded Entity Biographies2026EmoRES-TTS: Residual-Enhanced Vector Steering for Emotional Speech Generation2026Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering2026Traversing the solution space of neural networks with Hessian Null Space Continuation2026DataMagic: Authoring Data Videos through Declarative Multi-Agent Orchestration2026SpatialSpeak: QA-Native Reconstruction with Local and Global Context for Spatial Chain-of-Thought Reasoning2026Learning Dynamics of Continual Learning: A Unified View of Data Attribution, Forgetting, and Plasticity Loss2026StoryEngine: A State-Grounded Agentic Framework for Video Storytelling2026Precise Editing and Flexible Referencing for Interactable Worlds2026SEAD: A State-Based Perspective on Attack and Defense in Tool-Using Agents2026YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality2026KVCMAS: Efficient KV cache Correction for Shared Context in Multi-Agent Systems2026EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making2026LoopVL: Recurrent Visual Intelligence2026Training LLM Judges from Language Feedback via Position-Selective Self-Distillation2026A Missing Piece for Trustworthy AI Reviewers: From Benchmarking Rhetorical Robustness to SciCore Review2026RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement2026DAGent: Evaluate-then-Grow Planning for Deep Research Agents2026Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost2026Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior2026AutoDataBench: A Data-centric Testbed for Accelerating Auto Research2026How Much Is an AI Token Worth? Scaling Laws for Wild AI-Generated Web Text2026EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery2026Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL2026Initialization Improves LLM-Driven Discovery2026ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization2026Scaling and Distilling Text Embeddings for Better Diffusibility2026Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems2026Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes2026Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows2026From Knowledge Access to Source Learning: Developing Source-Specific Competence2026Hierarchical Continuous Diffusion Language Models2026ScholarCatalyst: A Benchmark for Retrieving Papers That Inspire New Research2026KaliBench: A Fine-Grained Benchmark for Cybersecurity Tool Use on Kali Linux with Runtime-Free Verifiable Rewards2026OmniConfess: Eliciting Token Confessions to Mitigate Omni-Modal Hallucination2026Gains and Collapse in On-Policy Distillation:A Reinforcement Learning Perspective2026Multilingual GSM-Symbolic: What determines capability transfer across languages?2026World Embedding Benchmark2026FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution2026Language Models that Play Chess and Explain Their Moves2026Periscope: Extending Frozen Language Models Beyond Their Context Window2026IdeaScientist: Orchestrating Agents for Grounded Scientific Ideation2026SEER: Self-Evolving Event Reasoning and Retrieval for Time Series Forecasting2026Questioning the Questions: Sustaining Self-Evolution in Reasoning Models2026SpecFold: Folding Multi-Branch Redundancy for Faster Speculative Decoding in Diffusion Language Models2026Self-Generated Feedback Destabilizes Test-Time Training: A Causal Decomposition of Long-Horizon Adaptation2026SearchJev: A Calibrated System-1 Model for Search Agents2026HLA: Expressive Hybrid Linear Attention via Chunk-Wise Dynamic Mixing2026Learning to Learn a Language2026Noise Out, Bias In: Targeted Bias Injection in Diffusion Language Models via Closed-Loop Activation Steering2026HuatuoGPT-3: RL-Only Domain Adaptation from Base Models2026From Traces to Agentic Worlds: Agentic Language World Models for Interactive Environment Simulation2026Judged Useless, Queried Anyway: Tool-Using Agents Rarely Turn Their Own Evidence Judgments into Stopping Decisions2026What Did the Agent Actually Do? Evidence-Grounded Oversight for Long-Horizon Agents2026The Assistance Dilemma: Learning to Teach via Multi-Turn Reinforcement Learning2026LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches2026Representation-Space MMD for Diffusion Language Models2026Base Models Can Reason By Taking a Cue From Training Data2026RMA: Context-Orchestrated Research Math Agents2026Beyond the Remembered World: Predictive 4D Belief for Persistent Navigation in Evolving Worlds2026Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers2026Video Generation Models: A Survey of Post-Training and Alignment2026GUI-HARVEST: Self-Improving GUI Agents through Evidence-Driven Harness Evolution2026VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks2026Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States2026Sharpening Tax in Post-Training2026Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents2026Local Support Learning2026DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation2026EditHero: A Benchmark for Long-Horizon Part-Level 3D Editing and Vibe Modeling2026DeskForge: Dense Supervision from Desktop Environments for Computer-Use Agents2026Inherit-MAS: Test-Time Evolution of Multi-Agent Systems through Workflow and Execution Inheritance2026Batched Speech Decisions Without Decoding: Single-Token Supervision Lets a Frozen LLM Hear Beyond the Transcript2026MetaRubric: Learning to Reward for Rubric-Based Reinforcement Learning2026