mlsanityData to 14 Oct 2026

Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning

arXiv 2507.00748 ↗First posted 1 Jul 2025cs.CVcited by 11

On the map: LLM reasoning → Multimodal reasoning

also in 1 more
Abstract (arXiv)

Multimodal Large Language Models (MLLMs) perform well in single-image visual grounding but struggle with real-world tasks that demand cross-image reasoning and multi-modal instructions. To address this, we adopt a reinforcement learning (RL) based post-training strategy for MLLMs in multi-image grounding tasks. We first synthesize high-quality chain-of-thought (CoT) data for cold-start initialization, followed by supervised fine-tuning (SFT) using low-rank adaptation (LoRA). Subsequently, we apply rejection sampling with the merged SFT model to curate reliable RL data and use rule-based RL to guide the model toward optimal reasoning paths. Extensive experiments demonstrate the effectiveness of our approach, achieving +9.04% on MIG-Bench and +4.41% on average across seven out-of-domain benchmarks.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 49 of its 53 references that are in the map. First posted on arXiv 1 Jul 2025.

Builds on

49 of 53 references.

Closest ancestors · 13

Same direction, newest first

  1. 2025TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning13 Apr
  2. 2025Perception-R1: Pioneering Perception Policy with Reinforcement Learning10 Apr
  3. 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model10 Apr
  4. 2025Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought8 Apr
  5. 2025Video-R1: Reinforcing Video Reasoning in MLLMs27 Mar
  6. 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization13 Mar
  7. 2025Boosting the Generalization and Reasoning of Vision Language Models with Curriculum Reinforcement Learning10 Mar
  8. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models9 Mar
  9. 2025Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement9 Mar
  10. 2025Visual-RFT: Visual Reinforcement Fine-Tuning3 Mar
  11. 2024LLaVA-CoT: Let Vision Language Models Reason Step-by-Step15 Nov
  12. 2024VISA: Reasoning Video Object Segmentation via Large Language Models16 Jul
1 more
  1. 2023LISA: Reasoning Segmentation via Large Language Model1 Aug

Other references · 36

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  5. 2023Visual Instruction Tuningcited by 7,869
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  7. 2025Qwen2.5-VL Technical Reportcited by 5,587
  8. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  9. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  10. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  11. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,834
  12. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
24 more
  1. 2021Grounded Language-Image Pre-trainingcited by 1,156
  2. 2024MiniCPM-V: A GPT-4V Level MLLM on Your Phonecited by 1,079
  3. 2016Modeling Context in Referring Expressionscited by 1,063
  4. 2015Generation and Comprehension of Unambiguous Object Descriptionscited by 904
  5. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  6. 2024LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Modelscited by 584
  7. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 550
  8. 2023GLaMM: Pixel Grounding Large Multimodal Modelcited by 455
  9. 2025R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimizationcited by 301
  10. 2024mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Modelscited by 291
  11. 2024MANTIS: Interleaved Multi-Image Instruction Tuningcited by 216
  12. 2025Time-R1: Post-Training Large Vision Language Model for Temporal Video Groundingcited by 146
  13. 2025Competitive Programming with Large Reasoning Modelscited by 119
  14. 2024GroundingGPT:Language Enhanced Multi-modal Grounding Modelcited by 101
  15. 2023V3Det: Vast Vocabulary Visual Detection Datasetcited by 81
  16. 2025Vision-R1: Evolving Human-Free Alignment in Large Vision-Language Models via Vision-Guided Reinforcement Learningcited by 69
  17. 2024MMIU: Multimodal Multi-image Understanding for Evaluating Large Vision-Language Modelscited by 67
  18. 2024Revisiting Referring Expression Comprehension Evaluation in the Era of Large Multimodal Modelscited by 53
  19. 2025Improved Visual-Spatial Reasoning via R1-Zero-Like Trainingcited by 51
  20. 2024MC-Bench: A Benchmark for Multi-Context Visual Grounding in the Era of MLLMscited by 39
  21. 2024Griffon v2: Advancing Multimodal Perception with High-Resolution Scaling and Visual-Language Co-Referringcited by 29
  22. 2024Lumen: Unleashing Versatile Vision-Centric Capabilities of Large Multimodal Modelscited by 24
  23. 2025Migician: Revealing the Magic of Free-Form Multi-Image Grounding in Multimodal Large Language Modelscited by 22
  24. 2024Advancing Fine-Grained Visual Understanding with Multi-Scale Alignment in Multi-Modal Modelscited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 6.

11 papers.

  1. 2025ChartMaster: Advancing Chart-to-Code Generation with Real-World Charts and Chart Similarity Reinforcement Learningcited by 16
  2. 2025Reinforcing Video Reasoning Segmentation to Think Before It Segmentscited by 15
  3. 2025Reinforcement Learning for Large Model: A Surveycited by 9
  4. 2025Reassessing the Role of Supervised Fine-Tuning: An Empirical Study in VLM Reasoningcited by 5
  5. 2026Don't Show Pixels, Show Cues: Unlocking Visual Tool Reasoning in Language Models via Perception Programscited by 4
  6. 2025PreferThinker: Reasoning-based Personalized Image Preference Assessmentcited by 0
  7. 2026RCoT-Seg: Reinforced Chain-of-Thought for Video Reasoning and Segmentationcited by 0
  8. 2026CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoningcited by 0
  9. 2026From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Modelscited by 0
  10. 2026VlogReward: Learning Multi-Dimensional Evaluation for Vlog Editingcited by 0
  11. 2026AffectOmni: RL-Verifiable People-Centric Grounded Affective Reasoning for Social and Art-Related Scenescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learning
  2. 2026GeM-VG: Towards Generalized Multi-image Visual Grounding with Multimodal Large Language Models
  3. 2025MIRG-RL: Multi-Image Reasoning and Grounding with Reinforcement Learning
  4. 2025High-Resolution Visual Reasoning via Multi-Turn Grounding-Based Reinforcement Learning
  5. 2024Plug-and-Play Grounding of Reasoning in Multimodal Large Language Models
  6. 2026iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning
  7. 2026PlaM: Training-Free Plateau-Guided Model Merging for Better Visual Grounding in MLLMs
  8. 2025Ground-R1: Incentivizing Grounded Visual Reasoning via Reinforcement Learning