CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal

The recent introduction of FigEx2, a visual-conditioned framework, aims to enhance the understanding of scientific compound figures by localizing panels and generating detailed captions directly from the images. This addresses the common issue of missing or inadequate captions that hinder panel-level comprehension.

A recent study has revealed that the group-relative advantage estimator used in Reinforcement Learning from Verifier Rewards (RLVR) is biased, systematically underestimating advantages for difficult prompts while overestimating them for easier ones. This imbalance can lead to ineffective exploration and exploitation strategies in training large language models.

The introduction of Process Relative Policy Optimization (PRPO) aims to enhance policy optimization for large language models (LLMs) by aligning process rewards with outcome rewards, addressing the limitations of existing critic-free methods like GRPO. PRPO provides a more nuanced approach by segmenting reasoning sequences and normalizing feedback, which improves the accuracy of models such as Qwen2.5-Math-1.5B on tasks like MATH500.