@inproceedings{wang-etal-2025-svd-llm,
    title = "{SVD}-{LLM} V2: Optimizing Singular Value Truncation for Large Language Model Compression",
    author = "Wang, Xin  and
      Alam, Samiul  and
      Wan, Zhongwei  and
      Shen, Hui  and
      Zhang, Mi",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2025.naacl-long.217/",
    pages = "4287--4296",
    ISBN = "979-8-89176-189-6",
    abstract = "Despite significant advancements, the practical deployment of Large Language Models (LLMs) is often hampered by their immense sizes, highlighting the need for effective compression techniques. Singular Value Decomposition (SVD) emerges as a promising method for compressing LLMs. However, existing SVD-based compression approaches suffer from substantial truncation losses, leading to severe performance degradation in compressed models. In this work, we introduce , a novel SVD-based LLM compression method that optimizes singular value truncation in SVD compression with two key strategies. First, employs dynamic compression ratio allocation to effectively balance the extremely large truncation loss across different layers. Second, it implements loss-optimized weight truncation to ensure that the truncated singular values result in a lower and more stable truncation loss in practice. We evaluate on ten datasets and five models on various scales and demonstrated that outperforms current state-of-the-art methods. The source code is available at \url{https://github.com/AIoT-MLSys-Lab/SVD-LLM}."
}