@inproceedings{li-etal-2025-faedkv,
    title = "{FAEDKV}: Infinite-Window {F}ourier Transform for Unbiased {KV} Cache Compression",
    author = "Li, Runchao  and
      Fu, Yao  and
      Sheng, Mu  and
      Long, Xianxuan  and
      Yu, Haotian  and
      Li, Pan",
    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.914/",
    doi = "10.18653/v1/2025.findings-emnlp.914",
    pages = "16856--16866",
    ISBN = "979-8-89176-335-7",
    abstract = "The efficacy of Large Language Models (LLMs) in long-context tasks is often hampered by the substantial memory footprint and computational demands of the Key-Value (KV) cache. Current compression strategies, including token eviction and learned projections, frequently lead to biased representations{---}either by overemphasizing recent/high-attention tokens or by repeatedly degrading information from earlier context{---}and may require costly model retraining. We present FAEDKV (Frequency-Adaptive Infinite-Window for KV cache), a novel, training-free KV cache compression framework that ensures unbiased information retention. FAEDKV operates by transforming the KV cache into the frequency domain using a proposed Infinite-Window Fourier Transform (IWDFT). This approach allows for the equalized contribution of all tokens to the compressed representation, effectively preserving both early and recent contextual information. A preliminary frequency ablation study identifies critical spectral components for layer-wise, targeted compression. Experiments on LongBench benchmark demonstrate FAEDKV{'}s superiority over existing methods by up to 22{\%}. In addition, our method shows superior, position-agnostic retrieval accuracy on the Needle-In-A-Haystack task compared to compression based approaches."
}