| GenClaw: Code-Driven Agentic Image Generation | Image & video gen | Junyan Ye +6 | May 28, 2026 | 38 |
| Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents | Agents | Suji Kim +2 | May 27, 2026 | 38 |
| LongAV-Compass: Towards Unified Evaluation of Minute-Scale Audio-Visual Generation Across T2AV, I2AV, and V2AV | Speech & audio | Tengfei Liu +19 | May 25, 2026 | 38 |
| Bridging Interleaved Multi-Modal Reasoning as a Unified Decision Process | Image & video gen | Zican Hu +12 | Jul 4, 2026 | 37 |
| GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation | Robotics | GigaWorld Team +26 | Jul 2, 2026 | 37 |
| Multi-Resolution Flow Matching: Training-Free Diffusion Acceleration via Staged Sampling | Image & video gen | Xingyu Zheng +6 | Jul 2, 2026 | 37 |
| CLI-Universe: Towards Verifiable Task Synthesis Engine for Terminal Agents | Agents | Zhanbo Hua +16 | Jun 22, 2026 | 37 |
| Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning | Agents | Xuekang Wang +5 | Jun 3, 2026 | 37 |
| Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer | Speech & audio | Ke Lei +6 | May 29, 2026 | 37 |
| Rethinking RAG in Long Videos: What to Retrieve and How to Use It? | Data & benchmarks | Yuho Lee +7 | Jun 11, 2026 | 36 |
| dMoE: dLLMs with Learnable Block Experts | Image & video gen | Sicheng Feng +4 | May 29, 2026 | 36 |
| SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer | Image & video gen | Yuyang Zhao +8 | May 28, 2026 | 36 |
| Wan-Streamer v0.2: Higher Resolution, Same Latency | Speech & audio | Lianghua Huang +25 | Jul 5, 2026 | 35 |
| AgenticDataBench: A Comprehensive Benchmark for Data Agents | Agents | Zhaoyan Sun +12 | Jul 2, 2026 | 35 |
| TurboServe: Serving Streaming Video Generation Efficiently and Economically | Image & video gen | Youhe Jiang +7 | Jun 17, 2026 | 35 |
| JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree Drafting | Image & video gen | Lanxiang Hu +11 | Jun 25, 2026 | 35 |
| MVTrack4Gen: Multi-View Point Tracking as Geometric Supervision for 4D Video Generation | Image & video gen | JoungBin Lee +10 | Jun 24, 2026 | 35 |
| From AGI to ASI | Agents | Tim Genewein +13 | Jun 10, 2026 | 35 |
| VIA-SD: Verification via Intra-Model Routing for Speculative Decoding | Code | Yuchen Xian +3 | Jun 10, 2026 | 35 |
| Human Psychometric Questionnaires Mischaracterize LLM Behavior | Safety & alignment | Woojung Song +5 | May 29, 2026 | 35 |
| VideoKR: Towards Knowledge- and Reasoning-Intensive Video Understanding | Vision & multimodal | Lin Fu +5 | Jun 3, 2026 | 35 |
| Qwen-Image-Flash: Beyond Objective Design | Image & video gen | Tianhe Wu +23 | Jun 2, 2026 | 35 |
| X-Stream: Exploring MLLMs as Multiplexers for Multi-Stream Understanding | Vision & multimodal | Peiwen Sun +11 | Jun 1, 2026 | 35 |
| SkillAdaptor: Self-Adapting Skills for LLM Agents from Trajectories | Agents | Zhuoyun Yu +6 | May 31, 2026 | 35 |
| NITP: Next Implicit Token Prediction for LLM Pre-training | Efficiency & systems | Xiangdong Zhang +5 | May 24, 2026 | 35 |
| ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence | Agents | Rui Meng +12 | May 25, 2026 | 35 |
| GEAR: Guided End-to-End AutoRegression for Image Synthesis | Image & video gen | Bin Lin +10 | Jun 30, 2026 | 34 |
| DeNovoSWE: Scaling Long-Horizon Environments for Generating Entire Repositories from Scratch | Agents | Jiale Zhao +6 | Jun 9, 2026 | 34 |
| LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Training | Vision & multimodal | Minju Gwak +5 | May 28, 2026 | 34 |
| Rethinking Memory as Continuously Evolving Connectivity | Agents | Jizhan Fang +14 | May 27, 2026 | 34 |
| AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction | Agents | Shanhui Zhao +15 | Jun 22, 2026 | 33 |
| Who Should Lead Decoding Now? Tracking Reliable Trajectories for Ensembling Masked Diffusion Language Models | Image & video gen | Heecheol Yun +3 | Jun 15, 2026 | 33 |
| Draft-OPD: On-Policy Distillation for Speculative Draft Models | Code | Haodi Lei +10 | May 28, 2026 | 33 |
| Native Audio-Visual Alignment for Generation | Speech & audio | Longbin Ji +8 | May 28, 2026 | 33 |
| Soap2Soap: Long Cinematic Video Remaking via Multi-Agent Collaboration | Image & video gen | Yiren Song +4 | May 17, 2026 | 33 |
| CUA-Gym: Scaling Verifiable Training Environments and Tasks for Computer-Use Agents | Agents | Bowen Wang +13 | May 25, 2026 | 33 |
| Infinite Worlds with Versatile Interactions | Image & video gen | Zelin Gao +19 | Jul 8, 2026 | 32 |
| WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory | Image & video gen | Hanlin Wang +12 | Jul 2, 2026 | 32 |
| EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory | Agents | Chang Nie +3 | Jun 19, 2026 | 32 |
| DF3DV-1K: A Large-Scale Dataset and Benchmark for Distractor-Free Novel View Synthesis | Image & video gen | Cheng-You Lu +8 | Jun 18, 2026 | 32 |
| From 2D Grids to 1D Tokens: Reforming Shared Representations for Multimodal Image Fusion | Vision & multimodal | Yuchen Xian +3 | Jun 10, 2026 | 32 |
| Lip Forcing: Few-Step Autoregressive Diffusion for Real-time Lip Synchronization | Speech & audio | Paul Hyunbin Cho +9 | Jun 9, 2026 | 32 |
| Rethinking the Divergence Regularization in LLM RL | Reinforcement learning | Jiarui Yao +5 | Jun 8, 2026 | 32 |
| Echo-Memory: A Controlled Study of Memory in Action World Models | Image & video gen | Wayne King +15 | Jun 8, 2026 | 32 |
| EarlyTom: Early Token Compression Completes Fast Video Understanding | Vision & multimodal | Hesong Wang +6 | May 28, 2026 | 32 |
| Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning | Agents | Jiapeng Zhu +7 | May 27, 2026 | 32 |
| Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems | Agents | Jianing Zhu +7 | May 25, 2026 | 32 |
| JLT: Clean-Latent Prediction in Latent Diffusion Transformers | Image & video gen | Funing Fu +4 | May 26, 2026 | 32 |
| Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation | Science & bio | Yifan Zhou +16 | Jul 9, 2026 | 31 |
| DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation | Code | Xin Cheng +32 | Jul 6, 2026 | 31 |
| ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving | Efficiency & systems | Sangjin Choi +5 | Jul 1, 2026 | 31 |
| OmniVideo-100K: A Dataset for Audio-Visual Reasoning through Structured Scripts and Evidence Chains | Speech & audio | Xinyue Cai +4 | Jun 12, 2026 | 31 |
| OVO-S-Bench: A Hierarchical Benchmark for Streaming Spatial Intelligence in Multimodal LLMs | Robotics | Yifei Li +6 | Jun 2, 2026 | 31 |
| Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios | Speech & audio | Changhao Pan +14 | May 27, 2026 | 31 |
| Share More, Search Less: Collaborative Parallel Thinking for Efficient Test-Time Scaling | Speech & audio | Xinglin Wang +11 | May 26, 2026 | 31 |
| VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon | Robotics | Yi Pan +10 | Jul 2, 2026 | 30 |
| Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity | Vision & multimodal | Bytedance Seed | Jun 30, 2026 | 30 |
| AsyncOPD: How Stale Can On-Policy Distillation Be? | Reinforcement learning | Wonjun Kang +11 | Jun 23, 2026 | 30 |
| Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models | Robotics | Haoqi Yuan +22 | Jun 17, 2026 | 30 |
| TRIAGE: Dialectical Reasoning for Explainable Risk Prediction on Irregularly Sampled Medical Time Series with LLMs | Code | Hyeongwon Jang +5 | Jun 8, 2026 | 30 |
| Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning | Vision & multimodal | Chaofan Ma +7 | Jun 10, 2026 | 30 |
| WorldOlympiad: Can Your World Model Survive a Triathlon? | Robotics | Yuke Zhao +10 | Jun 9, 2026 | 30 |
| Where to Look: Can Foundation Models Reach a Target Viewpoint Through Active Exploration? | Robotics | Liyang Li +7 | May 31, 2026 | 30 |
| SkillOpt-Lite: Better and Faster Agent Self-evolution via One Line of Vibe | Agents | Yifei Shen +2 | Jul 3, 2026 | 29 |
| MemSyco-Bench: Benchmarking Sycophancy in Agent Memory | Agents | Zhishang Xiang +7 | Jul 1, 2026 | 29 |
| AnchorWorld: Embodied Egocentric World Simulation with View-based Evolution Customization | Robotics | Yu Li +10 | Jun 5, 2026 | 29 |
| Streaming Communication in Multi-Agent Reasoning | Agents | Zhen Yang +5 | Jun 3, 2026 | 29 |
| World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning | Image & video gen | Yucheng Zhou +3 | Jun 2, 2026 | 29 |
| Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories | Reinforcement learning | Ali Behrouz +2 | Jun 2, 2026 | 29 |
| VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization | Image & video gen | Junhao Cheng +6 | Jun 1, 2026 | 29 |
| Linear Ensembles Wash Away Watermarks: On the Fragility of Distributional Perturbations in LLMs | Vision & multimodal | Zhihao Wu +4 | May 28, 2026 | 29 |
| LongE2V: Long-Horizon Event-based Video Reconstruction, Prediction, and Frame Interpolation with Video Diffusion Models | Image & video gen | Cheng-De Fan +6 | Jul 9, 2026 | 28 |
| UniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks | Vision & multimodal | Zhekai Chen +6 | Jul 9, 2026 | 28 |
| Multi-Turn Agentic Scientific Literature Search via Workflow Induction | Agents | Jisen Li +12 | Jul 1, 2026 | 28 |
| Breaking Failure Cascades: Step-Aware Reinforcement Learning for Medical Multimodal Reasoning | Vision & multimodal | Junha Jung +7 | Jun 30, 2026 | 28 |
| Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction | Vision & multimodal | Sunqi Fan +4 | Jun 28, 2026 | 28 |
| Fast LeWorldModel | Image & video gen | Yuntian Gao +1 | Jun 24, 2026 | 28 |
| GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents | Agents | Xiao Zhou +6 | Jun 22, 2026 | 28 |
| UnityShots: Memory-Driven Multi-Shot Audio-Video Generation with Boundary-Aware Gating | Speech & audio | Jiehui Huang +9 | Jun 19, 2026 | 28 |
| FreeStyle: Free Control of Style-Content Dual-Reference Generation from Community LoRA Mining | Vision & multimodal | Jinghong Lan +12 | Jun 18, 2026 | 28 |
| Guava: An Effective and Universal Harness for Embodied Manipulation | Robotics | Haowen Liu +7 | Jun 16, 2026 | 28 |
| VisualClaw: A Real-Time, Personalized Agent for the Physical World | Vision & multimodal | Haoqin Tu +16 | Jun 15, 2026 | 28 |
| Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation | Robotics | Jie Zhang +37 | Jun 15, 2026 | 28 |
| HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers | Vision & multimodal | Guozhen Zhang +13 | Jun 11, 2026 | 28 |
| Echo-Infinity: Learning Evolving Memory for Real-Time Infinite Video Generation | Image & video gen | Yuxuan Bian +11 | Jun 3, 2026 | 28 |
| AutoLab: Can Frontier Models Solve Long-Horizon Auto Research and Engineering Tasks? | Agents | Zhangchen Xu +18 | Jun 3, 2026 | 28 |
| SCOPE: Self-Play via Co-Evolving Policies for Open-Ended Tasks | Data & benchmarks | Wai-Chung Kwan +3 | May 29, 2026 | 28 |
| LLM-as-a-Tutor: Policy-Aware Prompt Adaptation for Non-Verifiable RL | Reinforcement learning | Yujin Kim +9 | Jul 5, 2026 | 27 |
| Multimodal Continuous Reasoning via Asymmetric Mutual Variational Learning | Vision & multimodal | Shijie Li +9 | Jul 1, 2026 | 27 |
| MemLearner: Learning to Query Context memory for Video World Models | Image & video gen | Jiwen Yu +9 | Jun 30, 2026 | 27 |
| DataEvolver: Self-Evolving Multi-Agent Data Construction for Text-Rich Image Generation | Image & video gen | Siyu Yan +4 | Jun 30, 2026 | 27 |
| PhotoQuilt: Training-Free Arbitrary-Resolution Photomosaics via Bootstrapped Tiled Denoising | Image & video gen | Koorosh Roohi +3 | Jun 29, 2026 | 27 |
| Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System | Robotics | Jiazhao Zhang +33 | Jun 18, 2026 | 27 |
| V-Zero: Answer-Label-Free On-Policy Distillation with Contrastive Evidence Gating for Fine-Grained Visual Reasoning | Vision & multimodal | Haoxiang Sun +8 | Jun 24, 2026 | 27 |
| OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation | Agents | Guibin Zhang +6 | Jun 16, 2026 | 27 |
| BRDFusion: Physics Meets Generation for Urban Scene Inverse Rendering | Science & bio | Yi-Ruei Liu +4 | Jun 15, 2026 | 27 |
| EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery | Agents | Amy Xin +7 | Jun 11, 2026 | 27 |
| Robots Need More than VLA and World Models | Robotics | Elis Karcini +8 | Jun 4, 2026 | 27 |
| OpenSkill: Open-World Self-Evolution for LLM Agents | Agents | Zhiling Yan +10 | Jun 4, 2026 | 27 |
| A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL | Code | Lei Yang +2 | Jun 1, 2026 | 27 |