一作ACL 2026Poster· Natural Language Processing
GRASP rune: Global Gating for Budgeted Structured Pruning of Large Language Models
Ziyang Wang · Jiangfeng Xiao · Chuan Xiao · Ruoxiang Li · Rui Mao · Jianbin Qin
摘要
Large language models (LLMs) are expensive to serve because dense FFN blocks, multi-head attention, and KV caches dominate memory, making structured pruning a natural way to reduce serving costs under tight parameter and memory budgets. We present GRASPrune, a global budgeted structured pruning f...