PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference.

Krishna Teja Chitty-Venkata, Jie Ye, Siddhisanket Raskar, Anthony Kougkas, Xian-He Sun, Murali Emani, Venkatram Vishwanath, Bogdan Nicolae

Browse the full EACL paper archive.