研究人员开发了MuCRASP，一种新颖的结构化剪枝框架，旨在减小视觉语言模型（VLM）的尺寸，同时不牺牲其链式思考（CoT）推理能力。现有的剪枝方法在处理VLM时存在困难，因为它们对CoT不敏感，并且没有考虑到跨模态激活差异。MuCRASP通过针对推理关键组件和保持跨模态对齐来解决这些问题，在压缩率和推理质量的基准测试中均显示出显著的改进。 AI
影响 通过减小复杂视觉语言模型的尺寸而不损害其推理能力，从而实现更高效的部署。
排序理由 该集群包含一篇学术论文，详细介绍了一种用于剪枝视觉语言模型的新方法。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.25842v1 Announce Type: new 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,…
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…