VRPRM: Process Reward Modeling via Visual Reasoning

arXiv:2508.03556 · cs.LG · Submitted 2025-08-05 · Read on arXiv

cs.LG

Submitted: 2025-08-05

Updated: 2026-10-16

Comments: 17 pages, 9 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Process Reward Model — 2× in this paper
• Chain-of-Thought (CoT) — 1× in this paper · explained in 5 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Process Reward Model (PRM) — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes
• Two-Stage Training Strategy — 1× in this paper · explained in 4 episodes
• Two-stage training — 1× in this paper · explained in 2 episodes

Sources

• RM-R1: Reward Modeling as Reasoning
• Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
• SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
• MM-PRM: Enhancing Multimodal Mathematical Reasoning with Scalable Step-Level Supervision
• Reward Reasoning Model
• Think-RM: Enabling Long-Horizon Reasoning in Generative Reward Models
• We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• HybridFlow: A Flexible and Efficient RLHF Framework
• WorldPM: Scaling Human Preference Modeling
• VisualPRM: An Effective Process Reward Model for Multimodal Reasoning
• Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning
• Unified Reward Model for Multimodal Understanding and Generation
• J1: Incentivizing Thinking in LLM-as-a-Judge via Reinforcement Learning
• LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts
• Generative Verifiers: Reward Modeling as Next-Token Prediction
• R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning
• The Lessons of Developing Process Reward Models in Mathematical Reasoning
• GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks