arXiv 2025 ยท preprint

Diagnosing visual reasoning: Challenges, insights, and a path forward

Figure: Diagnosing visual reasoning: Challenges, insights, and a path forward

Jing Bi, Guangyu Sun, Ali Vosoughi, Chen Chen, Chenliang Xu

A diagnosis of where multimodal models fail at visual reasoning, with directions for fixing it.

A diagnosis of where multimodal models fail at visual reasoning, with directions for fixing it.

All publications