SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation
Open Access
- 1 January 2025
- conference paper
- Published by Association for Computational Linguistics (ACL)
- p. 10775-10790
Abstract
Jialong Wu, Zhenglin Wang, Linhai Zhang, Yilong Lai, Yulan He, Deyu Zhou. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.Keywords
All Related Versions
This publication has 5 references indexed in Scilit:
- ClusterKV: Manipulating LLM KV Cache in Semantic Space for Recallable Compression2025
- LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding2024
- PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference2024
- ∞Bench: Extending Long Context Evaluation Beyond 100K Tokens2024
- H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models2023