| 2026-11-20 | 2611.26629 | Posterior Matching Distillation | Hidir Yesiltepe et al. | arXiv | 31 | |
| 2026-11-20 | 2611.23835 | NarrowSpec: Draft Hidden Width Need Not Match the Target in Speculative Decoding | Alagappan Valliappan et al. | arXiv | 3 | |
| 2026-11-13 | 2611.17331 | TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models | Chin-Yang Lin et al. | arXiv | 22 | |
| 2026-11-10 | 2611.14815 | BirdDreams: Joint Perspective and BEV Video Generation for a Plausible, Reactive Driving World Model | Junyoung Seo et al. | arXiv | 50 | |
| 2026-11-09 | 2610.35959 | AgentBaton: Efficient Model Collaboration in Agentic Tasks | Junlong Li et al. | arXiv | 19 | |
| 2026-11-09 | 2611.08290 | AutoCompress: Automated Hardware-Guided Model Compression | Sheldon Liang et al. | arXiv | 11 | |
| 2026-11-09 | 2611.09891 | CPU-ARC: Agentic Reasoning Caching for on-CPU Reasoning Reuse | Peter Belcak et al. | arXiv | 10 | |
| 2026-11-05 | 2611.08449 | VFlash: Fast, Accurate Visual Planning via Flash Primitive Decoding | Yu-Cheng Chou et al. | arXiv | 100 | |
| 2026-11-04 | 2611.05306 | You don't need long context | Peter Belcak et al. | arXiv | 14 | |
| 2026-11-04 | 2611.03998 | Dory1.0-4B: A Looped Normalized-GPT | Maxim Fishman et al. | arXiv | 11 | |
| 2026-10-27 | 2610.37666 | ThinkWorld: World Imagination Reinforces Chain-of-Dynamics Reasoning | Chaehyun Kim et al. | arXiv | 64 |  |
| 2026-10-27 | 2610.36466 | Nemotron-Labs-Diffusion-Uni: Elastic Self-Refining Unified Diffusion LLM with Cross-Paradigm On-Policy Distillation | Shufan Li et al. | arXiv | 30 | |
| 2026-10-26 | 2610.31779 | FLORA: An Open Model Suite for Scaling Robotics Data Engines | Sicong Leng et al. | arXiv | 3 | |
| 2026-10-23 | 2610.30340 | Foundation4D: Towards a unified perception and reconstruction model for autonomous vehicles | NVIDIA et al. | arXiv | 14 | |
| 2026-10-21 | 2610.28632 | Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary Rigs | Noam Rotstein et al. | arXiv | 27 | |
| 2026-10-19 | 2610.07898 | FC-SWE: Failure-Conditioned RL for Long-Horizon Software Engineering Agents | Jia Liufu et al. | arXiv | 8 | |
| 2026-10-16 | 2610.17560 | EdgeWise: Bringing Long-Horizon Agents to the Edge via Selective Cloud Delegation | Zeru Shi et al. | arXiv | 30 | |
| 2026-10-14 | 2610.18596 | ACCRUE: Continual Off-Policy Reinforcement Learning for Robot Foundation Models | Changyeon Kim et al. | arXiv | 20 | |
| 2026-10-14 | 2609.38377 | PhyProbe: Rethinking Physical Consistency Evaluation in Generated Videos | Max Ku et al. | arXiv | 15 | |
| 2026-10-14 | 2610.17540 | TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency | Jack Lu et al. | arXiv | 4 | |
| 2026-10-12 | 2610.11291 | When Do We Need On-Policy Distillation? Distilling on Offline Student Rollouts Is Often Better | Siyan Zhao et al. | arXiv | 28 | |
| 2026-10-12 | 2610.12582 | One Model to Loop Them All: Elastic Latent Recurrence for Language Models | Faisal Ladhak et al. | arXiv | 16 | |
| 2026-10-12 | 2610.11215 | GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images | Qirui Wu et al. | arXiv | 2 | |
| 2026-10-09 | 2606.03335 | A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning | Rui Zhang et al. | arXiv | 22 | |
| 2026-10-09 | 2610.11322 | USDCraft: Geometrically Grounded Programmatic Modeling of Articulated 3D Assets for Simulation | Chuanrui Zhang et al. | arXiv | 17 | |
| 2026-10-09 | 2610.09146 | Frozen Models, Evolving Expertise: Model-Agnostic Learning from Deployment Experience for Multimodal Medical AI | Yexiao He et al. | arXiv | 3 | |
| 2026-10-08 | 2610.10528 | Long-WAM: Scaling the Context of World-Action Models | Wei Huang et al. | arXiv | 97 | |
| 2026-10-07 | 2610.08430 | NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter Scale | Songlin Jiang et al. | arXiv | 28 | |
| 2026-10-07 | 2610.08463 | UNREAL: Unifying Retrieval and Long-Context with a Single Model | Edan Kinderman et al. | arXiv | 25 | |
| 2026-10-07 | 2610.08761 | VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning | Zewei Zhou et al. | arXiv | 9 | |
| 2026-10-06 | 2610.06647 | LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches | Shaokun Zhang et al. | arXiv | 45 | |
| 2026-10-06 | 2609.24890 | OSWorld-Pro: Process-based Evaluation for Computer Use Agents | Zhilin Wang et al. | arXiv | 23 | |
| 2026-10-06 | 2610.05750 | Beyond Semantic Similarity: Performance and Costs of Agentic Retrieval for Complex Tasks | Reza Esfandiarpoor et al. | arXiv | 17 | |
| 2026-10-01 | 2609.39982 | Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents | Minki Kang et al. | arXiv | 95 | |
| 2026-10-01 | 2609.38597 | PixelUMM: Encoder-Free Unified Image and Video Understanding and Generation | Cong Wei et al. | arXiv | 33 | |
| 2026-10-01 | 2609.40285 | PivotOPD: Learning to Recover from Pivotal Mistakes in Multi-Turn Agents | Yinghui He et al. | arXiv | 18 | |
| 2026-10-01 | 2609.40358 | Physis-Lang: Self-Evolving Language as a Physical Representation for Video World Model | Liming Lu et al. | arXiv | 16 | |
| 2026-09-30 | 2609.38154 | LongLive-Plug: Once-for-All Distillation for Video Generation | Shuai Yang et al. | arXiv | 29 | |
| 2026-09-30 | 2609.36864 | Where the Model Changes Its Mind: Hindsight-Divergence Localization for Efficient Reinforcement Learning with Verifiable Rewards | Fanchao Chen et al. | arXiv | 7 | |
| 2026-09-29 | 2609.35505 | An RL View of OPD: Least Square Policy Distillation for Sample-Efficient LLM Reasoning | Shangzhe Li et al. | arXiv | 28 | |
| 2026-09-29 | 2609.33419 | TT-VidT: Decoupling the Temporal Axis for Efficient Motion-Centric Video Pretraining | Shih-Ying Yeh et al. | arXiv | 18 | |
| 2026-09-18 | 2609.20519 | SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness | Haozhe Liu et al. | arXiv | 112 | |
| 2026-09-17 | 2609.18094 | Agora: Git as Shared Memory for Collective AutoResearch | Yifan Zhang et al. | arXiv | 35 | |
| 2026-09-11 | 2609.10712 | An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics | Ivan Moshkov et al. | arXiv | 21 | |
| 2026-09-09 | 2609.07108 | Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training | Zili Wang et al. | arXiv | 17 | |
| 2026-09-09 | 2609.07941 | ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding | Chia-Hui Chen et al. | arXiv | 4 | |
| 2026-09-03 | 2609.02849 | Post-Training Language Models for Gold-Medal Performance in Coding Competitions | Aleksander Ficek et al. | arXiv | 12 | |
| 2026-08-24 | 2608.18077 | Hydra-0: Action Flow for Generalist World Modeling and Control | Hongyu Li et al. | arXiv | 8 | |
| 2026-08-17 | 2608.10835 | UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations | Dvir Samuel et al. | arXiv | 4 | |
| 2026-08-14 | 2608.13391 | Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation | Hmrishav Bandyopadhyay et al. | arXiv | 11 | |
| 2026-08-10 | 2608.07408 | Addressable Memory for Video World Models | Xindi Wu et al. | arXiv | 15 | |
| 2026-07-30 | 2607.26410 | Voice Memory for Agentic Speech Recognition | Chao-Han Huck Yang et al. | arXiv | 10 | |
| 2026-07-29 | 2607.26004 | Parallel Decoding Distillation for Fast Image and Video Generation | Neta Shaul et al. | arXiv | 16 | |
| 2026-07-28 | 2607.24027 | Sol-Attn: Accelerating Video Generation Inference via On-the-Fly Attention Sparsification | Haopeng Li et al. | arXiv | 38 | |
| 2026-07-27 | 2607.21653 | Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement Learning | Jian Hu et al. | arXiv | 33 | |
| 2026-07-24 | 2607.21553 | SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation | Junsong Chen et al. | arXiv | 38 | |
| 2026-07-24 | 2607.20709 | NVIDIA-labs OO Agents: Native Python Object-Oriented Agents | Paul Furgale et al. | arXiv | 35 | |
| 2026-07-20 | 2607.16107 | Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex Videos | Sreyan Ghosh et al. | arXiv | 9 | |
| 2026-07-17 | 2607.15275 | RoboTTT: Context Scaling for Robot Policies | Yunfan Jiang et al. | arXiv | 17 | |
| 2026-07-10 | 2607.07740 | Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE | Haozhan Tang et al. | arXiv | 24 | |
| 2026-07-10 | 2607.08741 | ARDY: Autoregressive Diffusion with Hybrid Representation for Interactive Human Motion Generation | Kaifeng Zhao et al. | arXiv | 10 | |
| 2026-07-09 | 2607.06838 | WildCity: A Real-World City-Scale Testbed for Rendering, Simulation, and Spatial Intelligence | Xiangyu Han et al. | arXiv | 13 | |
| 2026-07-08 | 2607.03509 | Flex-Forcing: Towards a Unified Autoregressive and Bidirectional Video Diffusion Model | Xinyin Ma et al. | arXiv | 14 | |
| 2026-07-08 | 2607.05722 | Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding | Yonggan Fu et al. | arXiv | 11 | |
| 2026-07-07 | 2607.05196 | Unified Audio Intelligence Without Regressing on Text Intelligence | Zhifeng Kong et al. | arXiv | 21 | |
| 2026-07-02 | 2607.00272 | ASPIRE: Agentic /Skills Discovery for Robotics | Runyu Lu et al. | arXiv | 24 | |
| 2026-06-30 | 2606.29814 | Nemotron-Labs-Diffusion-Image: Advancing Masked Discrete Diffusion for High-Resolution Image Synthesis | Shufan Li et al. | arXiv | 14 | |
| 2026-06-30 | 2606.25621 | One Model, Many Latencies: Universal Speech Enhancement for Diverse Real-Time Applications | Szu-Wei Fu et al. | arXiv | 5 | |
| 2026-06-29 | 2606.28128 | PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation | Peiwen Zhang et al. | arXiv | 54 | |
| 2026-06-29 | 2606.28276 | SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation | Nadun Ranawaka et al. | arXiv | 16 | |
| 2026-06-29 | 2606.20905 | Vesta: A Generalist Embodied Reasoning Model | Johan Bjorck et al. | arXiv | 13 | |
| 2026-06-25 | 2606.25473 | Causal-rCM: A Unified Teacher-Forcing and Self-Forcing Open Recipe for Autoregressive Diffusion Distillation in Streaming Video Generation and Interactive World Models | Kaiwen Zheng et al. | arXiv | 27 | |
| 2026-06-19 | 2606.19980 | ENPIRE: Agentic Robot Policy Self-Improvement in the Real World | Wenli Xiao et al. | arXiv | 15 | |
| 2026-06-19 | 2606.18231 | Adaptive Volumetric Mechanical Property Fields Invariant to Resolution | Rishit Dagli et al. | arXiv | 3 | |
| 2026-06-17 | 2606.18216 | Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients | Byung-Kwan Lee et al. | arXiv | 64 | |
| 2026-06-17 | 2606.17321 | ProCUA-SFT Technical Report | Jaehun Jung et al. | arXiv | 10 | |
| 2026-06-16 | 2606.15007 | Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning | NVIDIA et al. | arXiv | 19 | |
| 2026-06-12 | 2606.13673 | SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning | Seokju Cho et al. | arXiv | 115 | |
| 2026-06-11 | 2606.04511 | SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference | Yaosheng Fu et al. | arXiv | 5 | |
| 2026-06-09 | 2606.07723 | VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation | Siyi Chen et al. | arXiv | 6 | |
| 2026-06-04 | 2606.02800 | Cosmos 3: Omnimodal World Models for Physical AI | Aditi et al. | arXiv | 142 | |
| 2026-06-04 | 2606.05160 | GRAIL: Generating Humanoid Loco-Manipulation from 3D Assets and Video Priors | Tianyi Xie et al. | arXiv | 9 | |
| 2026-06-03 | 2606.03159 | NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation | NVIDIA et al. | arXiv | 21 | |
| 2026-06-03 | 2606.03911 | Bootstrap Your Generator: Unpaired Visual Editing with Flow Matching | Yoad Tewel et al. | arXiv | 19 | |
| 2026-06-02 | 2606.02553 | LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generation | Qixin Hu et al. | arXiv | 17 | |
| 2026-06-02 | 2605.29318 | FreeForm: Reduced-Order Deformable Simulation from Particle-Based Skinning Eigenmodes | Donglai Xiang et al. | arXiv | 2 | |
| 2026-06-01 | 2605.30409 | SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer | Yuyang Zhao et al. | arXiv | 38 | |
| 2026-06-01 | 2605.31158 | Light Interaction: Training-Free Inference Acceleration for Interactive Video World Models | Jiacheng Lu et al. | arXiv | 3 | |
| 2026-05-29 | 2605.30161 | Why Far Looks Up: Probing Spatial Representation in Vision-Language Models | Cheolhong Min et al. | arXiv | 53 | |
| 2026-05-28 | 2605.28816 | Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players | Fangfu Liu et al. | arXiv | 95 | |
| 2026-05-28 | 2605.28774 | Agent Explorative Policy Optimization for Multimodal Agentic Reasoning | Minki Kang et al. | arXiv | 89 | |
| 2026-05-28 | 2605.23163 | Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving | Kewei Zhang et al. | arXiv | 16 | |
| 2026-05-27 | 2605.27365 | LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding | Shihao Wang et al. | arXiv | 142 | |
| 2026-05-25 | 2605.23902 | PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion | Yifan Lu et al. | arXiv | 45 | |
| 2026-05-22 | 2605.22791 | Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention | Ali Hatamizadeh et al. | arXiv | 28 | |
| 2026-05-20 | 2605.19786 | Fast 4D Mesh Generation by Spatio-Temporal Attention Chains | Dvir Samuel et al. | arXiv | 11 | |
| 2026-05-19 | 2605.18739 | LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation | Yukang Chen et al. | arXiv | 116 | |
| 2026-05-15 | 2605.15178 | SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer | Haoyi Zhu et al. | arXiv | 88 | |
| 2026-05-15 | 2605.14906 | MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models | Xiyu Ren et al. | arXiv | 77 | |
| 2026-05-14 | 2605.13724 | AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation | Yuchao Gu et al. | arXiv | 101 | |
| 2026-05-14 | 2605.05806 | Retrieval from Within: An Intrinsic Capability of Attention-Based Models | Elad Hoffer et al. | arXiv | 10 | |
| 2026-05-01 | 2604.24954 | Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence | NVIDIA et al. | arXiv | 16 | |