cs.LGNov 17, 2026

Eyes in the Update? Learning Where Rewards Land in Multimodal Reasoning

Authors: Lixuan He, Tianzhe Chu, Shulin Tian, Shikang Zheng, Feng Chen, Yi Ma

Organizations: The University of Hong Kong · TranscEngram · Nanyang Technological University

Abstract

Multimodal post-training has moved from broad instruction tuning toward reinforcement learning for image and video reasoning, where outcome rewards can improve final-answer accuracy. Yet these rewards are assigned to complete generated trajectories, while visual evidence is used only at sparse and uneven positions inside the response. Recent work has begun to make multimodal RL more token-sensitive through grounding signals or external token-level weighting. We ask a more specific question: can the allocation of the same trajectory-level reward be inferred from the model's own multimodal computation, rather than from an auxiliary annotator, perception module, or process label? We propose Gaze-RL, a representation-driven policy optimization method that redistributes completion-level advantage through a latent reward landing field. The field is computed from internal cross-modal geometry and local predictive uncertainty, assigning larger update weight to positions that are both visually anchored and decision-bearing while remaining mean-one, so the total credit of each rollout is preserved and only its positionwise allocation changes. Gaze-RL requires no process supervision, auxiliary perception module, or architectural modification, and improves both Qwen2.5-VL and InternVL3.5 across a broad range of image and video benchmarks.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Structured Role-Aware Policy Optimization for Multimodal Reasoning

    May 8, 2026Bingqing Jiang, Difan ZouToken-Level Credit AssignmentMultimodal Reasoning

  2. Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR

    Mar 27, 2026Jinda Lu, Junkang Wu, Jinghan Li +6RL for VLMsMultimodal Reasoning

  3. Same Reward, Different Skills: When Multimodal RL Learns to Look

    Oct 1, 2026Haocun Ye, Xinlong Jiang, Qile Chen +6Reinforcement Learning with Verifiable RewardsVisual Grounding