mlsanityData to 7 Oct 2026

Video-R1: Reinforcing Video Reasoning in MLLMs

arXiv 2503.21776 ↗First posted 27 Mar 2025cs.CVcited by 464

On the map: LLM reasoning → Multimodal reasoning

also in 1 more
Abstract (arXiv)

Inspired by DeepSeek-R1's success in eliciting reasoning abilities through rule-based reinforcement learning (RL), we introduce Video-R1 as the first attempt to systematically explore the R1 paradigm for incentivizing video reasoning within multimodal large language models (MLLMs). However, directly applying RL training with the GRPO algorithm to video reasoning presents two primary challenges: (i) a lack of temporal modeling for video reasoning, and (ii) the scarcity of high-quality video-reasoning data. To address these issues, we first propose the T-GRPO algorithm, which encourages models to utilize temporal information in videos for reasoning. Additionally, instead of relying solely on video data, we incorporate high-quality image-reasoning data into the training process. We have constructed two datasets: Video-R1-CoT-165k for SFT cold start and Video-R1-260k for RL training, both comprising image and video data. Experimental results demonstrate that Video-R1 achieves significant improvements on video reasoning benchmarks such as VideoMMMU and VSI-Bench, as well as on general video benchmarks including MVBench and TempCompass, etc. Notably, Video-R1-7B attains a 37.1% accuracy on video spatial reasoning benchmark VSI-bench, surpassing the commercial proprietary model GPT-4o. All code, models, and data are released in: https://github.com/tulerfeng/Video-R1.

Key facts. Cited by 464 papers in the map (data through 7 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 37 of its 51 references that are in the map. First posted on arXiv 27 Mar 2025.

Builds on

37 of 51 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models9 Mar
  2. 2024VISA: Reasoning Video Object Segmentation via Large Language Models16 Jul
  3. 2023Multimodal Chain-of-Thought Reasoning in Language Models2 Feb

Other references · 34

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  4. 2025Qwen2.5-VL Technical Reportcited by 5,418
  5. 2024GPT-4o System Cardcited by 4,467
  6. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  7. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  8. 2024Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysiscited by 1,375
  9. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,193
  10. 2023MVBench: A Comprehensive Multi-modal Video Understanding Benchmarkcited by 1,086
  11. 2024VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMscited by 749
  12. 2023VILA: On Pre-training for Visual Language Modelscited by 686
22 more
  1. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 655
  2. 2024Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spacescited by 580
  3. 2023LLaMA-VID: An Image is Worth 2 Tokens in Large Language Modelscited by 539
  4. 2024Long Context Transfer from Language to Visioncited by 502
  5. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 381
  6. 2024TempCompass: Do Video LLMs Really Understand Videos?cited by 352
  7. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 319
  8. 2025Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learningcited by 260
  9. 2025Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videoscited by 225
  10. 2024VideoLLM-online: Online Video Large Language Model for Streaming Videocited by 216
  11. 2023Video Understanding with Large Language Models: A Surveycited by 203
  12. 2024Advancing LLM Reasoning Generalists with Preference Treescited by 191
  13. 2025Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Modelscited by 144
  14. 2024Kangaroo: A Powerful Video-Language Model Supporting Long-context Video Inputcited by 132
  15. 2025MMVU: Measuring Expert-Level Multi-Discipline Video Understandingcited by 131
  16. 2024On Designing Effective RL Reward at Training Time for LLM Reasoningcited by 87
  17. 2025Fin-R1: A Large Language Model for Financial Reasoning through Reinforcement Learningcited by 82
  18. 2024Process Reward Model with Q-Value Rankingscited by 78
  19. 2023Audio-Visual LLM for Video Understandingcited by 76
  20. 2025Unhackable Temporal Rewarding for Scalable Video MLLMscited by 33
  21. 2024Keypoint-based Progressive Chain-of-Thought Distillation for LLMscited by 22
  22. 2025Will Pre-Training Ever End? A First Step Toward Next-Generation Foundation MLLMs via Self-Improving Systematic Cognitioncited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 220; 2026: 244.

464 papers.

Sort:

By area:

  1. 2025Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learningcited by 267
  2. 2025VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuningcited by 248
  3. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 175
  4. 2025SpaceR: Reinforcing MLLMs in Video Spatial Reasoningcited by 152
  5. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planningcited by 152
  6. 2025Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawingcited by 130
  7. 2025Perception-R1: Pioneering Perception Policy with Reinforcement Learningcited by 116
  8. 2025Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?cited by 107
  9. 2025A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibilitycited by 89
  10. 2025Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Modelscited by 89
  11. 2025Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoningcited by 88
  12. 2025VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuningcited by 86
38 more
  1. 2025Visionary-R1: Mitigating Shortcuts in Visual Reasoning with Reinforcement Learningcited by 86
  2. 2025Scaling RL to Long Videoscited by 83
  3. 2025SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Modelscited by 83
  4. 2025Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoningcited by 67
  5. 2025R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learningcited by 65
  6. 2025Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiencycited by 65
  7. 2025Think Only When You Need with Large Hybrid-Reasoning Modelscited by 64
  8. 2025GUI-G2: Gaussian Reward Modeling for GUI Groundingcited by 60
  9. 2025VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perceptioncited by 60
  10. 2025TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoningcited by 59
  11. 2025Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Viewscited by 58
  12. 2025LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Callingcited by 58
  13. 2025DeepVideo-R1: Video Reinforcement Fine-Tuning via Difficulty-aware Regressive GRPOcited by 57
  14. 2025Look-Back: Implicit Visual Re-focusing in MLLM Reasoningcited by 54
  15. 2025OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLMcited by 53
  16. 2025Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Visioncited by 52
  17. 2025OneThinker: All-in-one Reasoning Model for Image and Videocited by 52
  18. 2025ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understandingcited by 50
  19. 2025HumanOmniV2: From Understanding to Omni-Modal Reasoning with Contextcited by 50
  20. 2025GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoningcited by 46
  21. 2026Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understandingcited by 46
  22. 2025SophiaVL-R1: Reinforcing MLLMs Reasoning with Thinking Rewardcited by 45
  23. 2025Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Modelscited by 43
  24. 2025Open-o3-Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidencecited by 43
  25. 2025VisPlay: Self-Evolving Vision-Language Models from Imagescited by 42
  26. 2025ACTIVE-o3: Empowering MLLMs with Active Perception via Pure Reinforcement Learningcited by 41
  27. 2025Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmarkcited by 41
  28. 2025FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlightingcited by 39
  29. 2025Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaborationcited by 38
  30. 2025Reinforced Visual Perception with Toolscited by 38
  31. 2025VideoScore2: Think before You Score in Generative Video Evaluationcited by 37
  32. 2025Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulationcited by 36
  33. 2025Harnessing the Reasoning Economy: A Survey of Efficient Reasoning for Large Language Modelscited by 35
  34. 2025Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecyclecited by 34
  35. 2025TUNA: Taming Unified Visual Representations for Native Unified Multimodal Modelscited by 34
  36. 2025OctoNav: Towards Generalist Embodied Navigationcited by 33
  37. 2025VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understandingcited by 32
  38. 2025Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learningcited by 32

50 most cited of 464 All 464 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Exploring the Effect of Reinforcement Learning on Video Understanding: Insights from SEED-Bench-R1
  2. 2026Video-ToC: Video Tree-of-Cue Reasoning
  3. 2025VerIPO: Cultivating Long Reasoning in Video-LLMs via Verifier-Gudied Iterative Policy Optimization
  4. 2025VideoSeg-R1:Reasoning Video Object Segmentation via Reinforcement Learning
  5. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model
  6. 2026Video Models Can Reason with Verifiable Rewards
  7. 2026Video-KTR: Reinforcing Video Reasoning via Key Token Attribution
  8. 2025SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models