Eigen Attention: Attention in Low-Rank Space for KV Cache Compression

TL;DR

Eigen Attention introduces a low-rank space approach to compress KV caches in large language models, significantly reducing memory and latency during inference with minimal performance loss.

Contribution

The paper presents Eigen Attention, a novel method that performs attention in a low-rank space to reduce KV cache size and latency, compatible with existing compression techniques.

Findings

Up to 40% reduction in KV cache size

Up to 60% reduction in attention latency

Minimal performance degradation

Abstract

Large language models (LLMs) represent a groundbreaking advancement in the domain of natural language processing due to their impressive reasoning abilities. Recently, there has been considerable interest in increasing the context lengths for these models to enhance their applicability to complex tasks. However, at long context lengths and large batch sizes, the key-value (KV) cache, which stores the attention keys and values, emerges as the new bottleneck in memory usage during inference. To address this, we propose Eigen Attention, which performs the attention operation in a low-rank space, thereby reducing the KV cache memory overhead. Our proposed approach is orthogonal to existing KV cache compression techniques and can be used synergistically with them. Through extensive experiments over OPT, MPT, and Llama model families, we demonstrate that Eigen Attention results in up to 40% reduction in KV cache sizes and up to 60% reduction in attention operation latency with minimal drop in performance. Code is available at https://github.com/UtkarshSaxena1/EigenAttn.