Physical Plausibility Reasoning via HCM-GRPO: Empowering Compact Model for Superior Performance
On the map: LLM reasoning → Multimodal reasoning
Abstract (arXiv)
The performance of image generation has been significantly improved in recent years. However, the study of image screening is rare, and its performance with Multimodal Large Language Models (MLLMs) is unsatisfactory due to the lack of data and the weak physical plausibility reasoning ability in MLLMs. In this work, we propose a complete solution to address these problems in terms of data and methodology. For data, we collect a comprehensive image screening dataset with over 128k samples, comprising about 640k images. Each sample consists of an original image and four generated images. The dataset evaluates the physical plausibility reasoning ability under four aspects: appearance deformation, physical shadow, placement layout, and extension rationality. Regarding data annotation, we investigate multiple approaches, including purely manual, fully automated, and answer-driven annotations, to acquire high-quality chains of thought (CoT) data in the most cost-effective manner. Methodologically, we introduce a Hard Cases Mining (HCM) strategy with a Dynamic Proportional Accuracy (DPA) reward into the Group Relative Policy Optimization (GRPO) framework, called HCM-GRPO. This enhanced method demonstrates superior physical plausibility reasoning capabilities compared to the original GRPO. Our experimental results reveal that even state-of-the-art closed-source MLLMs, such as GPT5.2 and Gemini3-Pro, exhibit unsatisfactory performance in physical plausibility reasoning. In contrast, by leveraging the HCM-GRPO, we are able to surpass the scores of both large-scale open-source and leading closed-source models with a much smaller model.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 25 of its 41 references that are in the map. First posted on arXiv 13 Nov 2025.
Builds on
25 of 41 references.
Closest ancestors · 5
Same direction, newest first
- 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency25 Aug
- 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning1 Jul
- 2025Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning6 May
- 2025Visual-RFT: Visual Reinforcement Fine-Tuning3 Mar
- 2024CoMT: A Novel Benchmark for Chain of Multi-modal Thought on Large Vision-Language Models17 Dec
Other references · 20
- 2020Denoising Diffusion Probabilistic Modelscited by 21,076
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
- 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,503
- 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
- 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
- 2023IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Modelscited by 1,464
- 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,023
- 2023AGIEval: A Human-Centric Benchmark for Evaluating Foundation Modelscited by 716
- 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 550
- 2023Generative Multimodal Models are In-Context Learnerscited by 446
- 2024SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuningcited by 355
- 2024OminiControl: Minimal and Universal Control for Diffusion Transformercited by 340
8 more
- 2025R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimizationcited by 301
- 2024MuirBench: A Comprehensive Benchmark for Robust Multi-image Understandingcited by 188
- 2024Bilateral Reference for High-Resolution Dichotomous Image Segmentationcited by 112
- 2024NaturalBench: Evaluating Vision-Language Models on Natural Adversarial Samplescited by 97
- 2024CC-OCR: A Comprehensive and Challenging OCR Benchmark for Evaluating Large Multimodal Models in Literacycited by 81
- 2024GalleryGPT: Analyzing Paintings with Large Multimodal Modelscited by 28
- 2024Enhancing textual textbook question answering with large language models and retrieval augmented generationcited by 14
- 2025Towards Fine-Grained Vision-Language Alignment for Few-Shot Anomaly Detectioncited by 4
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2025SPR-128K: A New Benchmark for Spatial Plausibility Reasoning with Multimodal Large Language Models
- 2025PixelThink: Towards Efficient Chain-of-Pixel Reasoning
- 2024Plug-and-Play Grounding of Reasoning in Multimodal Large Language Models
- 2026Faithful GRPO: Improving Visual Spatial Reasoning in Multimodal Language Models via Constrained Policy Optimization
- 2025Boosting MLLM Reasoning with Text-Debiased Hint-GRPO
- 2025DiffThinker: Towards Generative Multimodal Reasoning with Diffusion Models
- 2026Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design
- 2025Evaluating MLLMs with Multimodal Multi-image Reasoning Benchmark