2 papers
cs.CV2026
Reinforcement Learning with Robust Rubric Rewards
Ya-Qi Yu, Hao Wang, Fangyu Hong +15
While Reinforcement Learning with Verifiable Rewards (RLVR) is effective for deterministically checkable tasks, many vision-language tasks are partially verifiable, demanding multi…
cs.CV2026
Visual Preference Optimization with Rubric Rewards
Ya-Qi Yu, Fangyu Hong, Xiangyang Qu +15
The effectiveness of Direct Preference Optimization (DPO) depends on preference data that reflect the quality differences that matter in multimodal tasks. Existing pipelines often…