MuCRASP: Multimodal Chain-of-thought Reasoning aware Structured Pruning 事件

MuCRASP: Multimodal Chain-of-thought Reasoning aware Structured Pruning arXiv:2605.25842v1 Announce Type: cross Abstract: Vision-language models (VLMs) increasingly rely on chain-of-thought (CoT) reasoning to solve complex multimodal tasks, but their large parameter sizes make deployment expensive. Structured pruning offers a natural solution; however, existing methods fail to preserve CoT reasoning accuracy in VLMs. We identify two key reasons: (1) CoT consistency depends on sparse transition