cs.AI · 2606.07801 Copy arXiv ID · Jun 5, 2026 Save Improving Multimodal Reasoning via Worst Dimension Optimization Authors: Haocheng Lv , Huaping Zhang , Qiuchi Li , Lei Li , Chunxiao Gao
Organizations: Beijing Institute of Technology
Abstract Multimodal reasoning requires a path that retains integrity over a wide range of constraints, from visual grounding to logic consistency. However, the current Process Reward Models focus on heuristically defined rewards that equally weigh these factors, which may lead to the concealment of individual dimension failures by the dominating factors, without guaranteeing the validity of the reasoning process in general.
Explore similar work Apr 23, 2026 · Jingpei Wu, Xiao Han, Weixiang Shen +3 Multimodal Reasoning Thinking
Apr 20, 2026 · cs.CL J/K move · Enter open · S save
Mengzhao Jia, Zhihan Zhang, Meng Jiang
University of Notre Dame
Reinforcement Learning with Verifiable Rewards (RLVR) improves multimodal reasoning by rewarding verifiable final answers. Yet answer-correct trajectories may still rely on incomplete derivations, weak evidence, or statements that contradict their conclusions. This gap between answer correctness and reasoning validity, which we call reasoning-answer inconsistency, motivates trajectory supervision in multimodal RL. We compare two main approaches: reward models (RMs), and Generative Rewards (GRs). RMs are efficient and help early in training, but their gains weaken as the policy distribution shifts; GRs improve performance, but may give unstable rewards and computationally expensive. We therefore propose Groupwise Ranking Reward, which ranks verifier-passed trajectories for the same prompt in one pass and redistributes reward accordingly. Groupwise comparison better separates stronger and weaker correct trajectories with lower judge overhead than GRs. Experiments show that RLVR aggravates reasoning-answer inconsistency, while trajectory supervision alleviates it. Groupwise Ranking Reward performs best overall, improving reliability-conditioned accuracy from 47.4% to 54.7% over RLVR.