AI RESEARCH
CAVE: A Structured Credit Assignment Approach for Fragmented Visual Evidence Reasoning
arXiv CS.CV
•
ArXi:2605.16416v1 Announce Type: new Vision-Language Models (VLMs) have achieved strong performance on general multimodal reasoning, yet remain challenged in integrating nonlocal visual information to semantically underdetermined visual reasoning. We describe this challenge as Fragmented Visual Reasoning. To this end, we propose Credit Assignment for Visual Evidence (CAVE), a structured process-reward method based on GRPO for interleaved visual reasoning.