VRPRM: Process Reward Modeling via Visual Reasoning
This is experimental HTML to improve accessibility. We invite you to report rendering errors. Use Alt+Y to toggle on accessible reporting links and Alt+Shift+Y to toggle off. Learn more about this project and help improve conversions.
VRPRM: Process Reward Modeling via Visual Reasoning Xinquan Chen 1 , Bangwei Liu 1,2 , Xuhong Wang 1 , Yingchun Wang 1 , Chaochao Lu 1 Corresponding Author: wangxuhong@pjlab.org.cn Abstract Process Reward Model (PRM) is widely used in the post-training of Large Language Model (LLM) because it can perform fine-grained evaluation of the reasoning steps of generated content. However, most PRMs lack long-term reasoning and deep thinking capabilities. On the other hand, although a few works have tried to introduce Chain-of-Thought capability into PRMs, the annotation cost of CoT-PRM data is too exp
saved by
related reading
- MolmoAct Action Reasoning Models that can Reason in Spacearxiv.org
- Reasoning in General Domains without Verifiersarxiv.org
- Let's Verify Step by Steparxiv.org
- [2201.11903] Chain of Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
- State of RL for reasoning LLMs | A. Weersaweers.de
- Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-Experts | RLHFlowrlhflow.github.io
- RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedbackarxiv.org
- [2504.13837] Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?arxiv.org
- [2505.17373] Value-Guided Search for Efficient Chain-of-Thought Reasoningarxiv.org
- Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domainsarxiv.org
- [2506.19733] Breaking Barriers: Do Reinforcement Post Training Gains Transfer To Unseen Domains?arxiv.org
- Training Large Language Models to Reason in a Continuous Latent Spacearxiv.org