WeightedKV: Attention Scores Weighted Key-Value Cache Merging for Large Language Models.
Jian Yuan, Ziwei He, Haoli Bai, Jingwen Leng, Bo Jiang
Browse the full ICASSP paper archive.
Jian Yuan, Ziwei He, Haoli Bai, Jingwen Leng, Bo Jiang
Browse the full ICASSP paper archive.