Year: 2024
Type: preprint
Abstract: This paper introduces a novel approach, the Bounded-Cache Transformer (BCT), for building large language models with a predefined Key-Value (KV) cache capacity. The BCT addresses the excessive memory consumption issue in traditional KV caches by implementing a bounded-length KV cache, which is particularly suitable for the attention layers in Transformer decode-only architectures. By dynamically updating the key-value vector sequences, the BCT achieves efficient inference within limited cache capacity, significantly reducing memory usage while maintaining model performance and system throughput. Experimental results demonstrate that the BCT significantly reduces memory usage while maintaining the model's inference quality, offering a new solution for efficient inference in large language models.
Source: arXiv (Cornell University)
Authors: Yi, Zhonghua, Niu, Ge, Wang, Lei, Tang, Wei, Zhang, Liqiu
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4405023509