LSH Tells You What To Discard: An Adaptive Locality-Sensitive Strategy for KV Cache Compression

Tahseen Rabbani, Minghui Liu, Tony O'Halloran, Ananth Sankaralingam, Mary-Anne Hartley, Furong Huang

Keywords: kv cache, locality-sensitive hashing, compression

TL;DR: We develop a token eviction strategy that uses locality-sensitive hashing to locate low-attention tokens without computing attention.

Abstract: Transformer-based large language models (LLMs) use the key-value (KV) cache to significantly accelerate inference by storing the key and value embeddings of past tokens. However, this cache consumes significant GPU memory. In this work, we introduce LSH-E, an algorithm that uses locality-sensitive hashing (LSH) to compress the KV cache. LSH-E quickly locates tokens in the cache that are cosine dissimilar to the current query token. This is achieved by computing the Hamming distance between binarized Gaussian projections of the current token query and cached token keys, with a projection length much smaller than the embedding dimension. We maintain a lightweight binary structure in GPU memory to facilitate these calculations. Unlike existing compression strategies that compute attention to determine token retention, LSH-E makes these decisions pre-attention, thereby reducing computational costs. Additionally, LSH-E is dynamic -- at every decoding step, the key and value of the current token replace the embeddings of a token expected to produce the lowest attention score. We demonstrate that LSH-E can compress the KV cache by 30\%-70\% while maintaining high performance across reasoning, multiple-choice, long-context retrieval and summarization tasks.

Primary Area: foundation or frontier models, including LLMs

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 6181