MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding

ACL ARR 2024 June Submission1996 Authors

Abstract: Auto-regressive inference of transformers benefit greatly from Key-Value (KV) caching, but can lead to major memory bottlenecks as model size, batch size, and sequence length grow at scale. We introduce Multi-Layer Key-Value (MLKV) sharing, a novel approach extending KV sharing across transformer layers to reduce memory usage beyond what was possible with Multi-Query Attention (MQA) and Grouped-Query Attention (GQA). Evaluations on various NLP benchmarks and inference metrics using uptrained Pythia-160M variants demonstrate that MLKV significantly reduces memory usage with minimal performance loss, reducing KV cache size down to a factor of 6x compared to MQA. These results highlight MLKV's potential for efficient deployment of transformer models at scale.

Paper Type: Long

Research Area: Special Theme (conference specific)

Research Area Keywords: model architectures, efficient models, efficiency in model inference

Contribution Types: NLP engineering experiment, Approaches low compute settings-efficiency

Languages Studied: english

Submission Number: 1996