1
Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score
提出Diet-LLM方法，通过融合任务特定重要性分数实现维度级全局剪枝，在保持性能的同时显著压缩大模型规模。
arXiv:2603.23985v3 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated remarkable capabilities, but their massive scale po…