@inproceedings{wang-etal-2026-grasprune,
    title = "{GRASP}rune: Global Gating for Budgeted Structured Pruning of Large Language Models",
    author = "Wang, Ziyang  and
      Xiao, Jiangfeng  and
      Xiao, Chuan  and
      LI, Ruoxiang  and
      Mao, Rui  and
      Qin, Jianbin",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.491/",
    pages = "10719--10736",
    ISBN = "979-8-89176-390-6",
    abstract = "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 framework applied post-hoc to a pretrained model that jointly prunes FFN channels and attention KV head groups under a single global parameter budget. GRASPrune attaches lightweight learnable gates to prunable units and optimizes only these gates on a small unlabeled language-modeling calibration set, keeping all backbone weights frozen while enforcing the target sparsity at every step. A final budget-preserving scaling calibration reweights the surviving channels and heads to correct scale shifts introduced by pruning. On LLaMA-2-7B, GRASPrune removes 50{\%} of parameters and achieves 12.18 perplexity on WikiText-2 while maintaining competitive average zero-shot accuracy on five downstream benchmarks, using a short calibration run of four epochs on 512 unlabeled sequences on a single NVIDIA A100 80GB GPU, all without any full-model fine-tuning."
}