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.
FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation2026MemUse: Moving Memory Evaluation from Direct QA to Natural Integration in Long-Term Human-AI Conversation2026FrontierChallenge: Evaluating Scientific Workflow Completion2026DataKernelBench: Can LLMs Optimize Database Queries on GPUs?2026Paint What You See: Benchmarking Dexterous Visual Tool Use in Multimodal Agents2026CaSKG: Counterfactual-Causal Skill Graphs for Scalable Agent Skill Retrieval2026Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models2026LibriBrain100: One Hundred Hours of Broad and Deep MEG Data for Neural Speech Decoding at Scale2026JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolution2026Skill Issue: Are Skills Language-Invariant in LLMs?2026Prefix Sliding for efficient test-time scaling2026VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning2026GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models2026Code World Model: Coding Agent as World Brain2026V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement Learning2026GameWAM: A World Action Model for Video Games2026J-Zero: Unified Challenger--Solver--Judge Self-Evolution from Zero Data2026Generative Semantic Scene Completion2026Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $50902026TTPO: Test-Time Policy Optimization2026VoiceMem: Streaming Dual-Brain Memory for Real-Time Interaction2026AgentJudgeBench: A Multi-Difficulty Benchmark for Evaluating LLM Judges on Agentic Tool-Calling2026LeVJEPA: Efficient & Scalable Video Pretraining without the Heuristics2026PAWBench: How Far Are We from Probabilistically Aligned World Modeling?2026RealSWE: A Compositional Evaluation of Coding Agents under Realistic User Requests2026OpenStamp: A Watermark for Open-Source Language Models2026LoopArena: Benchmarking Models as Runtime Controllers for Loop Engineering2026EvoUndo: Recoverability-Constrained Self-Evolution for LLM Agent Harnesses2026Fast Weight Attention for Continual Learning2026ContextPilot: Teaching Agents for Proactive Context Management via Fine-grained RL2026Blind Men and the Elephant: Probing the Epistemic Myopia of LLMs under Long-Tail Divergent Knowledge2026Evaluating the Hidden Costs of Personalization in Large Language Models2026SafeAtlas-VL: Beyond Binary Multimodal Safety with Large-Scale Data and Guard Models2026Locked at the Entrance, Open Inside: Where RLVR Narrows the Solution Space2026Dynamic Important Example Mining for Reinforcement Finetuning2026QCell: Recombining and Aligning Cell Queries for Overlapping Instance Segmentation2026Super Library Agent: Joint Generation and Maintenance of Multiple Applications Beyond the Single Codebase2026EvoGenUI-Bench: Evaluating LLMs as Multi-Turn Generative UI Assistants2026Chain-of-Thought Faithfulness of Reasoning Models Varies with Where and How Preference Cues Are Delivered2026Cross-lingual Functional Vectors for Emotion Detection in Large Language Models2026E-Commerce Bench: Evaluating LLM Agents on Long-Horizon Autonomous Business Operation2026Normalized Low-Rank Adaptation2026Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement2026Learning to Evaluate Before Improving: Automatic Rubric Induction for Automatic Research Agents2026SpanCalib-VLM: Calibrated Hallucination Span Detection in Vision-Language Models2026On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability2026CogEvol: Towards Efficient and Reliable Learning Environment Generation2026PaperGym: Rubric-Centered Evolution for Research-Plan Generation2026LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation2026UI-Venus-2 Technical Report2026Safin-1: Safety from Within through Memory-Native State Evolution2026WHALE: A Simple Recipe for Joint Harness-Weight Optimization2026Dr. Claw: An AI Scientist Workspace for Vibe Research2026EM^2Mem: Event-Centric Multimodal Memory for Large Language Models2026Let Confidence Change, Not the Prediction: Prediction-Preserving Repair for Post-hoc Calibration2026GUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent Environments2026Enoki: Efficient Multi-Level Hallucination Detection2026Control-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs2026Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall2026StudentSim: Training LLM-based Student Simulators2026Scaling Large Reasoning Models beyond Human Supervision: A Path toward Superintelligence2026Recursive Criticality of AI Self-Improvement2026EmbodiedSkills: World Action Model-augmented Cascaded Skills for Vision-Language-Action Agent2026Evaluating Multimodal LLMs as Generalist Vision-Language-Action Agents for Drone Control: Commanding, Approaching, Tracking and Searching2026Harness-of-Harness: Multi-Day Autonomous Software Development with Continual Improvement2026Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens2026A Common Measure of Communication for Speech Brain-Computer Interfaces2026PaperCompiler: Faithful Paper-to-Code Generation via Repository-Level Specification Compilation2026Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills2026ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding2026EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction2026TempCloze: Can Video-LLMs Identify the Missing Middle?2026Unifying Conformal Language Tasks with In-Context Ensembles2026What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation2026Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning2026Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs2026Editable Visual Design2026When Models Edit Too Much: On the Fidelity of Minimal Code Edits2026CORE: Improving Compositional Reasoning in MLLM Embedding via Reranker Distillation2026Rethinking On-Policy Distillation of Large Language Models II: One Training Example2026Last Translation Benchmark2026Compile by Training: Turning Natural-Language Specifications into Local Neural Functions2026Causal Foundation Models2026H3-World: Turning Language Understanding into World Control2026FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning Experience2026Unlocking Lossless Speedups in LLMs via Discrete Diffusion2026DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training2026Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems2026The Attention Triangle in Audio-Video Models2026ENEAS: Embedding-guided Neural Ensemble for Adaptive Segmentation2026LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes2026One Editor, Many Edits: A Unified Training-Free Framework for Diverse Video Editing2026Iris: Climbing to the Search Frontier2026VLA-Precision: Asymmetric Co-Bootstrapping for Efficient Real-World Online RL of Vision-Language-Action Models2026MaxKernel: Agentic Kernel Generation for TPUs2026$τ^τ$-Bench: An Environment for End-To-End, Realistic Agent Construction2026Refuse without Refusal: A Structural Analysis of Safety-Tuning Responses for Reducing False Refusals in Language Models2026Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models2026Ask Before You Optimize: Dynamic Pre-Formulation Clarification for Interactive Optimization2026RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?2026Privacy Failure in Split-LLM Training, The Returned Gradient Nullifies the Decoys2026BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference2026UniMate: One Unified Model to Animate Diverse Skeletons2026Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs2026WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data2026SQS: Bayesian DNN Compression through Sparse Quantized Sub-distributions2025What LLM Trading Agents Actually Do in Production: A Six-Month, Population-Scale Record from Two Fleets2026Diffs vs. Whole Files: An Empirical Comparison of Iterative Edit-Based and Direct Generation for Flutter/Dart Code Models2026Steering Geometry: Validating Human Value Geometry in LLM Steering Space2026OracleZoom: On-Policy Self-Distillation Inspired Reference-Constrained Recursive Image Super Resolution2026Reason Through the Latent! Making Latent Visual Reasoning Necessary2026SpatialBlock: Enhancing Spatial Intelligence in LVLMs via Synthetic Block-Stacking Problem2026Ambient @ EgoProactive 2026 : Proactive Egocentric Assistance with Visually Grounded Supervision2026RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting2026Harnessing CLIP and DINO: An Uncertainty-Aware Cascaded Fusion Network for Generalizable Deepfake Image Detection2026Kalman Delta Networks: Uncertainty-aware Associative Memory2026A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM2026SchemeArena: Factorized Stress Testing of Scheming in LLM Agents2026SWE-Bench Pro Verified: A Reliable Benchmark for Software Engineering Agents2026Environments as Scaffold: Enriching Feedback to Bootstrap Self-Evolving Agents in Long-Horizon Tasks2026Difficulty-Adaptive Tree-Structured Policy Optimization for Expanding Reasoning Coverage in RLVR2026PlannerForge: LLM Agents for Scenario-Based Testing of Motion Planners in Autonomous Driving2026Multimodal Duplex Interaction Agent2026SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?2026SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models2026RenderFormer-V2: Neural Rendering with Heterogeneous Scene Primitives2026Broken on Arrival: Non-Functional Quantized LLM Artifacts in Public Model Registries and How to Catch Them2026Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model2026Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning2026MOLE: Detecting Insider Threats in AI Agents2026Continual Learning Mechanisms Compose for Long-Horizon Memorization2026Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training2026Marigold V2: Revisiting Diffusion Transformers for Monocular Depth Estimation2026ActionSplice: In-Flight Action Editing for Interactive World Models2026CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs2026Miles v0.1: Production-Level Post-Training2026Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation2026Grounded Skill Synthesis from Code at Scale for Agentic Intelligence2026PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents2026Revisiting Complete Reasoning Traces for Post-Training2026NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness2026AuK Technical Report: An Open-Source Foundational Model for Speech Generation and Editing2026ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation2026Studying Image Tokenizers as Visual Languages in Unified Multimodal Models2026Scores Alone Do Not Prove Discovery: The Discovery Certification Protocol for Auditing AI Research Agents2026RESCUE-BENCH: Towards Relation-Aware Multi-Party Emotional Support Conversation Systems2026MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes2026Reference-Based Bias Detection in LLMs via Relative Representations of Hidden States2026Show-Harness: Just a VLM Agent Can Play Robots2026$Φ$-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?2026Why Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs2026IdeaAMBIG: Benchmarking Implementation-Critical Gaps in Research-Idea Specifications2026HuRo: Robotizing Human Videos for Scalable VLA Pretraining2026NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction2026Think Before You Link: Rarity, Reasoning, and Retrieval in Multilingual Entity Linking2026Negative Self-Distillation: Learning to Reason by Avoiding Flaws2026The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement2026An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics2026Towards a Deterministic Math Solver for Clinical Language Models2026Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation2026X-AuT: Progressive Audio-Encoder Compression for Speech LLMs with Cross-Scale Distillation2026COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization2026Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation2026Expert-Space Exploration in MoE Reinforcement Learning2026SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking2026Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work2026Pelican-Sim 1.0: A General World Model Simulator for Embodied Intelligence2026UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation2026Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model2026ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search2026Thought without systematicity? Evaluating reasoning models on rule induction tasks2026Convergent Emergence of In-Context Learning Across Modalities2026Pick Your Poison: Learning to Select Poison Sets for Stronger LLM Backdoor Attacks2026HazardAuditor: From Executable Threats to Safer Computer-Use Agents2026RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments2026KaiNinja: Extending Native 3D Generators to the Part Level2026A Language-Guided Multimodal Foundation Model for Zero-Shot and Multi-Task Brain Signal Analysis2026EvoOntology: A Self-Evolving Ontology Layer for Data Agents2026Atria Dawn: The Dawn of Agentic Superintelligence2026JumpStart Your Policy Learning with Lessons from 160,000 Training Runs2026Training Specialist Models without Reasoning Trajectories for Domain Expert Distillation2026Omni-Streaming Thinking2026Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States2026Disentangling Representation Evolution in Transformers through Directional Decomposition2026E2A-Bench: Benchmarking Evidence-to-Action Reliability in Financial Chart Reasoning2026SpectralShift: Effective Context Window Extension of Gated DeltaNet via Spectral Reparameterization2026ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement2026Dream-RSI: Recursive Self-Improvement through Evolving Worlds2026When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis2026Discovery Foundation Models: Toward Open-Ended Discovery Intelligence2026OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learning2026CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design2026AI for Games in the Foundation Model Era2026LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence2026Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration2026ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals2026The Functionalizer: Lossless Functional Decomposition for Subword Tokenization2026Register Tokens for Bounded-State Reasoning in Diffusion Language Models2026RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation2026ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents2026