Segment-Aligned Policy Optimization for Multi-Modal Reasoning
A comprehensive benchmark specifically designed to evaluate the reasoning capability of MLLMs.
lei-gao
•
•
1 min read
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
System review of reward hacking.
xiaohua-wang
Omniquality-R: Advanced Reward Models through All-Encompassing Quality Assessment
A unified reward modeling framework that transforms multi-task quality reasoning into continuous and interpretable reward signals .
yiting-lu