KV-CAR: KV Cache Compression using Autoencoders and KV Reuse in Large Language Models
This addresses a critical memory efficiency problem for LLM inference, enabling longer sequences and larger batch sizes, though it is an incremental improvement on existing compression techniques.
The paper tackles the memory bottleneck of KV cache in large language models during autoregressive decoding by proposing KV-CAR, a framework combining autoencoder compression and similarity-driven reuse of KV tensors, achieving up to 47.85% KV cache memory reduction with minimal impact on perplexity and accuracy.
As Large Language Models (LLMs) scale in size and context length, the memory requirements of the key value (KV) cache have emerged as a major bottleneck during autoregressive decoding. The KV cache grows with sequence length and embedding dimension, often exceeding the memory footprint of the model itself and limiting achievable batch sizes and context windows. To address this challenge, we present KV CAR, a unified and architecture agnostic framework that significantly reduces KV cache storage while maintaining model fidelity. KV CAR combines two complementary techniques. First, a lightweight autoencoder learns compact representations of key and value tensors along the embedding dimension, compressing them before they are stored in the KV cache and restoring them upon retrieval. Second, a similarity driven reuse mechanism identifies opportunities to reuse KV tensors of specific attention heads across adjacent layers. Together, these methods reduce the dimensional and structural redundancy in KV tensors without requiring changes to the transformer architecture. Evaluations on GPT 2 and TinyLLaMA models across Wikitext, C4, PIQA, and Winogrande datasets demonstrate that KV CAR achieves up to 47.85 percent KV cache memory reduction with minimal impact on perplexity and zero shot accuracy. System level measurements on an NVIDIA A40 GPU show that the reduced KV footprint directly translates into longer sequence lengths and larger batch sizes during inference. These results highlight the effectiveness of KV CAR in enabling memory efficient LLM inference.