mlsanityData to 14 Oct 2026

Perception-R1: Pioneering Perception Policy with Reinforcement Learning

arXiv 2504.07954 ↗First posted 10 Apr 2025cs.CV cs.CLcited by 120

On the map: LLM reasoning → Multimodal reasoning

also in 1 more
Abstract (arXiv)

Inspired by the success of DeepSeek-R1, we explore the potential of rule-based reinforcement learning (RL) in MLLM post-training for perception policy learning. While promising, our initial experiments reveal that incorporating a thinking process through RL does not consistently lead to performance gains across all visual perception tasks. This leads us to delve into the essential role of RL in the context of visual perception. In this work, we return to the fundamentals and explore the effects of RL on different perception tasks. We observe that the perceptual complexity is a major factor in determining the effectiveness of RL. We also observe that reward design plays a crucial role in further approching the upper limit of model perception. To leverage these findings, we propose Perception-R1, a scalable RL framework using GRPO during MLLM post-training. With a standard Qwen2.5-VL-3B-Instruct, Perception-R1 achieves +4.2% on RefCOCO+, +17.9% on PixMo-Count, +4.2% on PageOCR, and notably, 31.9% AP on COCO2017 val for the first time, establishing a strong baseline for perception policy learning.

Key facts. Cited by 120 papers in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 57 of its 68 references that are in the map. First posted on arXiv 10 Apr 2025.

Builds on

57 of 68 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Video-R1: Reinforcing Video Reasoning in MLLMs27 Mar
  2. 2025Visual-RFT: Visual Reinforcement Fine-Tuning3 Mar
  3. 2023MMBench: Is Your Multi-modal Model an All-around Player?12 Jul

Other references · 54

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2022Training language models to follow instructions with human feedbackcited by 15,971
  6. 2017Proximal Policy Optimization Algorithmscited by 15,215
  7. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  8. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  9. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  10. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  11. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  12. 2023Visual Instruction Tuningcited by 7,869
38 more
  1. 2020End-to-End Object Detection with Transformerscited by 7,414
  2. 2017Mask R-CNNcited by 7,277
  3. 2025Qwen2.5-VL Technical Reportcited by 5,587
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  5. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  6. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  7. 2023Qwen Technical Reportcited by 3,344
  8. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  9. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  10. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  11. 2018YOLOv3: An Incremental Improvementcited by 2,659
  12. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
  13. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,385
  14. 2025s1: Simple test-time scalingcited by 1,341
  15. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,239
  16. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  17. 2016Modeling Context in Referring Expressionscited by 1,063
  18. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 914
  19. 2015Generation and Comprehension of Unambiguous Object Descriptionscited by 904
  20. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 838
  21. 2021MDETR -- Modulated Detection for End-to-End Multi-Modal Understandingcited by 819
  22. 2016A Diagram Is Worth A Dozen Imagescited by 804
  23. 2022OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Frameworkcited by 745
  24. 2023CogAgent: A Visual Language Model for GUI Agentscited by 704
  25. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  26. 2025Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Modelcited by 461
  27. 2023DreamLLM: Synergistic Multimodal Comprehension and Creationcited by 297
  28. 2023Nougat: Neural Optical Understanding for Academic Documentscited by 245
  29. 2024mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understandingcited by 220
  30. 2023Making LLaMA SEE and Draw with SEED Tokenizercited by 185
  31. 2024General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Modelcited by 180
  32. 2024DreamBench++: A Human-Aligned Benchmark for Personalized Image Generationcited by 131
  33. 2024mDPO: Conditional Preference Optimization for Multimodal Large Language Modelscited by 92
  34. 2023ChatSpot: Bootstrapping Multimodal LLMs via Precise Referring Instruction Tuningcited by 65
  35. 2024Interpretable Contrastive Monte Carlo Tree Search Reasoningcited by 64
  36. 2024Focus Anywhere for Fine-grained Multi-page Document Understandingcited by 62
  37. 2023Merlin:Empowering Multimodal LLMs with Foresight Mindscited by 46
  38. 2024Small Language Model Meets with Reinforced Vision Vocabularycited by 43

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 58; 2026: 62.

120 papers.

Sort:

By area:

  1. 2025Latent Visual Reasoningcited by 104
  2. 2025SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Modelscited by 90
  3. 2025Visual Planning: Let's Think Only with Imagescited by 85
  4. 2025UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learningcited by 84
  5. 2025VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforcement Learningcited by 68
  6. 2025OneThinker: All-in-one Reasoning Model for Image and Videocited by 56
  7. 2025Look-Back: Implicit Visual Re-focusing in MLLM Reasoningcited by 55
  8. 2025Ground-R1: Incentivizing Grounded Visual Reasoning via Reinforcement Learningcited by 49
  9. 2025Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Modelscited by 44
  10. 2025Advancing Multimodal Reasoning via Reinforcement Learning with Cold Startcited by 41
  11. 2025Patho-R1: A Multimodal Reinforcement Learning-Based Pathology Expert Reasonercited by 40
  12. 2025Spotlight on Token Perception for Multimodal Reinforcement Learningcited by 39
38 more
  1. 2025Jigsaw-R1: A Study of Rule-based Visual Reinforcement Learning with Jigsaw Puzzlescited by 36
  2. 2025Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoningcited by 34
  3. 2026STEP3-VL-10B Technical Reportcited by 32
  4. 2025One RL to See Them All: Visual Triple Unified Reinforcement Learningcited by 30
  5. 2025MoDoMoDo: Multi-Domain Data Mixtures for Multimodal LLM Reinforcement Learningcited by 30
  6. 2025AlphaOne: Reasoning Models Thinking Slow and Fast at Test Timecited by 29
  7. 2026LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decodingcited by 28
  8. 2025Rex-Thinker: Grounded Object Referring via Chain-of-Thought Reasoningcited by 26
  9. 2025Sketch-in-Latents: Eliciting Unified Reasoning in MLLMscited by 25
  10. 2025RePrompt: Reasoning-Augmented Reprompting for Text-to-Image Generation via Reinforcement Learningcited by 23
  11. 2025WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learningcited by 21
  12. 2025LENS: Learning to Segment Anything with Unified Reinforced Reasoningcited by 18
  13. 2025Perception Before Reasoning: Two-Stage Reinforcement Learning for Visual Reasoning in Vision-Language Modelscited by 17
  14. 2025Scaling Agentic Reinforcement Learning for Tool-Integrated Reasoning in VLMscited by 17
  15. 2025ChartMaster: Advancing Chart-to-Code Generation with Real-World Charts and Chart Similarity Reinforcement Learningcited by 16
  16. 2025SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agentscited by 15
  17. 2025TempSamp-R1: Effective Temporal Sampling with Reinforcement Fine-Tuning for Video LLMscited by 15
  18. 2025Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Modelscited by 14
  19. 2026Act Wisely: Cultivating Meta-Cognitive Tool Use in Agentic Multimodal Modelscited by 14
  20. 2025Retrv-R1: A Reasoning-Driven MLLM Framework for Universal and Efficient Multimodal Retrievalcited by 12
  21. 2026Thinking in Uncertainty: Mitigating Hallucinations in MLRMs with Latent Entropy-Aware Decodingcited by 12
  22. 2025PeRL: Permutation-Enhanced Reinforcement Learning for Interleaved Vision-Language Reasoningcited by 11
  23. 2025Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learningcited by 11
  24. 2025Agentic Jigsaw Interaction Learning for Enhancing Visual Perception and Reasoning in Vision-Language Modelscited by 11
  25. 2025TTRV: Test-Time Reinforcement Learning for Vision Language Modelscited by 10
  26. 2025SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Modelscited by 10
  27. 2025Video-R2: Reinforcing Consistent and Grounded Reasoning in Multimodal Language Modelscited by 10
  28. 2025ProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems in the Wildcited by 10
  29. 2026V-Zero: Self-Improving Multimodal Reasoning with Zero Annotationcited by 10
  30. 2026PhyCritic: Multimodal Critic Models for Physical AIcited by 10
  31. 2026PointRFT: Explicit Reinforcement Fine-tuning for Point Cloud Few-shot Learningcited by 10
  32. 2025Reinforcement Learning for Large Model: A Surveycited by 9
  33. 2025Perception-Consistency Multimodal Large Language Models Reasoning via Caption-Regularized Policy Optimizationcited by 9
  34. 2025CoPRS: Learning Positional Prior from Chain-of-Thought for Reasoning Segmentationcited by 9
  35. 2025VACoT: Rethinking Visual Data Augmentation with VLMscited by 9
  36. 2026CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoningcited by 9
  37. 2026Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thoughtcited by 9
  38. 2026CreatiParser: Generative Image Parsing of Raster Graphic Designs into Editable Layerscited by 9

50 most cited of 120 All 120 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model
  2. 2025G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning
  3. 2026On Asymmetric Optimization of Reasoning and Perception in Vision-Language Model Post-Training
  4. 2025Rethinking RL Scaling for Vision Language Models: A Transparent, From-Scratch Framework and Comprehensive Evaluation Scheme
  5. 2025Perception-R1: Advancing Multimodal Reasoning Capabilities of MLLMs via Visual Perception Reward
  6. 2025Vision-R1: Evolving Human-Free Alignment in Large Vision-Language Models via Vision-Guided Reinforcement Learning
  7. 2026Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning
  8. 2026All Roads Lead to Rome: Incentivizing Divergent Thinking in Vision-Language Models