@inproceedings{eyceoz-etal-2025-hopscotch,
    title = "Hopscotch: Discovering and Skipping Redundancies in Language Models",
    author = "Eyceoz, Mustafa  and
      Nayak, Nikhil Shivakumar  and
      Wang, Hao  and
      Han, Ligong  and
      Srivastava, Akash",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.861/",
    doi = "10.18653/v1/2025.findings-emnlp.861",
    pages = "15904--15913",
    ISBN = "979-8-89176-335-7",
    abstract = "Modern causal language models stack many attention blocks to improve performance, but not all blocks are necessary for every task. We propose Hopscotch, a simple yet effective method that identifies and skips attention blocks with least contributions to a task and adapts to preserve output quality. Hopscotch jointly optimizes which blocks to skip and how to scale the outputs of the remaining layers. By introducing lightweight, trainable scaling parameters to attention and MLP blocks, it mitigates distribution shifts in hidden states caused by removing attention blocks. Hopscotch does not modify model weights or require access to pretraining or instruction-tuning data, and is compatible with existing model compression techniques. When applied to $\texttt{Llama-3.1-8B}$ and $\texttt{Qwen-2.5-7B}$, Hopscotch achieves less than a 2{\%} drop in performance even after skipping four attention blocks."
}