Residual Visual Credit Optimization: Conserved Evidence Routing for Multimodal Reinforcement Learning

arXiv:2610.04918 · cs.LG, cs.CL · Submitted 2026-10-04 · Read on arXiv

cs.LG, cs.CL

Submitted: 2026-10-04

Updated: 2026-10-04

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

Terminology

• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• The Limited Multi-Label Projection Layer
• Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
• Qwen3-VL Technical Report
• Qwen2.5-VL Technical Report
• Learning Self-Correction in Vision-Language Models via Rollout Augmentation
• The Llama 3 Herd of Models
• Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models
• LLaVA-OneVision: Easy Visual Task Transfer
• PRPO: Perception-Reinforced Policy Optimization via Token-Level Dynamic Advantage Reshaping
• Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thought
• MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning
• Proximal Policy Optimization Algorithms
• VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning
• Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs
• VisNumBench: Evaluating Number Sense of Multimodal Large Language Models
• LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts
• Not All Tokens See Equally: Perception-Grounded Policy Optimization for Large Vision-Language Models
• DAPO: An Open-Source LLM Reinforcement Learning System at Scale

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