Rethinking Visual Dependency in Long-Context Reasoning for Large Vision-Language Models

LRM
Re-assign community