Co-occurring keywords · papers mentioning both
contrastive group relative policy optimization
(1)
reinforcement learning with verifiable reward
(1)
image quality evaluation
(1)
computer-use agent
(1)
reward reasoning
(1)
step-level filtering
(1)
multi-stage reinforcement learning
(1)
reweighting training samples
(1)
shortcut-aware learning
(1)
unimodal spurious correlations
(1)
Papers
MSRL: Scaling Generative Multimodal Reward Modeling via Multi-Stage Reinforcement Learning
CVPR 2026
Now They See It, Now They Don’t: Multimodal Reward Models Exhibit Unreliability in Physical World Constraints
CoNLL 2026
WebSTAR: Scalable Data Synthesis for Computer Use Agents with Step-Level Filtering
ACL 2026
EQA-RM: A Generative Embodied Reward Model with Test-time Scaling
EMNLP 2025
The Devil Is in the Details: Tackling Unimodal Spurious Correlations for Generalizable Multimodal Reward Models
ICML 2025