@inproceedings{eyal-etal-2025-layer,
    title = "Layer Duplication in {LLM}s",
    author = "Eyal, Neo  and
      Dershowitz, Nachum  and
      Bar, Kfir",
    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.967/",
    doi = "10.18653/v1/2025.findings-emnlp.967",
    pages = "17797--17807",
    ISBN = "979-8-89176-335-7",
    abstract = "We investigate the effect of duplicating multihead self-attention layers in large language models (LLMs) across a range of language tasks, with and without fine-tuning. The results demonstrate that duplicating the initial layers once or twice often yields a significant performance boost. Attention analysis uncovered the underlying mechanisms driving the improvement when performing layer duplication. This method enhances LLM capabilities with or without additional training or labeled data."
}