mlsanityData to 14 Oct 2026

CrowdVLM-R1: Expanding R1 Ability to Vision Language Model for Crowd Counting using Fuzzy Group Relative Policy Reward

arXiv 2504.03724 ↗First posted 31 Mar 2025cs.CV cs.CLcited by 15

On the map: LLM reasoning → Reinforcement learning for reasoning

also in 2 more
Abstract (arXiv)

We propose Fuzzy Group Relative Policy Reward (FGRPR), a novel framework that integrates Group Relative Policy Optimization (GRPO) with a fuzzy reward function to enhance learning efficiency. Unlike the conventional binary 0/1 accuracy reward, our fuzzy reward model provides nuanced incentives, encouraging more precise outputs. Experimental results demonstrate that GRPO with a standard 0/1 accuracy reward underperforms compared to supervised fine-tuning (SFT). In contrast, FGRPR, applied to Qwen2.5-VL(3B and 7B), surpasses all baseline models, including GPT4o, LLaMA2(90B), and SFT, across five in-domain datasets. On an out-of-domain dataset, FGRPR achieves performance comparable to SFT but excels when target values are larger, as its fuzzy reward function assigns higher rewards to closer approximations. This approach is broadly applicable to tasks where the precision of the answer is critical. Code and data: https://github.com/yeyimilk/CrowdVLM-R1

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 26 of its 48 references that are in the map. First posted on arXiv 31 Mar 2025.

Builds on

26 of 48 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization17 Mar
  2. 2017Proximal Policy Optimization Algorithms20 Jul

Other references · 24

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,332
  4. 2015VQA: Visual Question Answeringcited by 2,506
  5. 2025Visual-RFT: Visual Reinforcement Fine-Tuningcited by 513
  6. 2022Visual Classification via Description from Large Language Modelscited by 336
  7. 2013OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networkscited by 224
  8. 2025MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learningcited by 173
  9. 2018Context-Aware Crowd Countingcited by 150
  10. 2022From Images to Textual Prompts: Zero-shot VQA with Frozen Large Language Modelscited by 115
  11. 2017Switching Convolutional Neural Network for Crowd Countingcited by 93
  12. 2022CounTR: Transformer-based Generalised Visual Countingcited by 79
12 more
  1. 2025Vision-R1: Evolving Human-Free Alignment in Large Vision-Language Models via Vision-Guided Reinforcement Learningcited by 69
  2. 2022A Low-Shot Object Counting Network With Iterative Prototype Adaptationcited by 65
  3. 2022Few-shot Object Counting with Similarity-Aware Feature Enhancementcited by 62
  4. 2023CrowdCLIP: Unsupervised Crowd Counting via Vision-Language Modelcited by 47
  5. 2023VLCounter: Text-aware Visual Representation for Zero-Shot Object Countingcited by 45
  6. 2020Global Wheat Head Detection (GWHD) dataset: a large and diverse dataset of high resolution RGB labelled images to develop and benchmark wheat head detection methodscited by 42
  7. 2017CNN-based Cascaded Multi-task Learning of High-level Prior and Density Estimation for Crowd Countingcited by 41
  8. 2023Open-world Text-specified Object Countingcited by 41
  9. 2023T-Rex: Counting by Visual Promptingcited by 20
  10. 2024Benchmarking Large Language Models for Image Classification of Marine Mammalscited by 4
  11. 2022Mask Focal Loss: A unifying framework for dense crowd counting with canonical object detection networkscited by 3
  12. 2023Counting Manatee Aggregations using Deep Neural Networks and Anisotropic Gaussian Kernelcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 9; 2026: 6.

15 papers.

Sort:

By area:

  1. 2025Look-Back: Implicit Visual Re-focusing in MLLM Reasoningcited by 55
  2. 2025WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learningcited by 21
  3. 2025Sailing by the Stars: A Survey on Reward Models and Learning Strategies for Learning from Rewardscited by 14
  4. 2025Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Modelscited by 14
  5. 2025Object Detection with Multimodal Large Vision-Language Models: An In-depth Reviewcited by 14
  6. 2025ReDit: Reward Dithering for Improved LLM Policy Optimizationcited by 8
  7. 2026What does RL improve for Visual Reasoning? A Frankenstein-Style Analysiscited by 4
  8. 2026Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Designcited by 3
  9. 2026SITUATE -- Synthetic Object Counting Dataset for VLM trainingcited by 2
  10. 2025HOI-R1: Exploring the Potential of Multimodal Large Language Models for Human-Object Interaction Detectioncited by 1
  11. 2026Structure Over Scale: Learning Visual Reasoning from Pedagogical Videocited by 1
  12. 2026REC-RL: Referring expression counting via Gaussian and range-based reward optimizationcited by 1
3 more
  1. 2025Cross-Resolution SAR Target Detection Using Structural Hierarchy Adaptation and Reliable Adjacency Alignmentcited by 0
  2. 2025When Large Language Models Do Not Work: Online Incivility Prediction through Graph Neural Networkscited by 0
  3. 2026CrowdCue: Specialist-Cue Conditioning for Vision-Language Crowd Countingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Demystifying Group Relative Policy Optimization: Its Policy Gradient is a U-Statistic
  2. 2026Range-GRPO: Policy Optimization via Pairwise Relations among Reward Intervals
  3. 2025GVPO: Group Variance Policy Optimization for Large Language Model Post-Training
  4. 2026From Absolute to Relative: Rethinking Reward Shaping in Group-Based Reinforcement Learning
  5. 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
  6. 2025Token-Level Policy Optimization: Linking Group-Level Rewards to Token-Level Aggregation via Markov Likelihood
  7. 2026Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL
  8. 2026CoDistill-GRPO: A Co-Distillation Recipe for Efficient Group Relative Policy Optimization