{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T06:56:02Z","timestamp":1791874562867,"version":"4.3.4"},"reference-count":19,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10889583","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T13:52:43Z","timestamp":1741787563000},"page":"1-5","source":"Crossref","is-referenced-by-count":4,"title":["WeightedKV: Attention Scores Weighted Key-Value Cache Merging for Large Language Models"],"prefix":"10.1109","author":[{"given":"Jian","family":"Yuan","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziwei","family":"He","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haoli","family":"Bai","sequence":"additional","affiliation":[{"name":"Huawei Noah’s Ark Lab"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingwen","family":"Leng","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Jiang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref2","article-title":"Mistral 7b","author":"Jiang","year":"2023"},{"key":"ref3","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00632"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.307"},{"key":"ref6","article-title":"Code llama: Open foundation models for code","author":"Roziere","year":"2023"},{"key":"ref7","article-title":"Efficiently scaling transformer inference. corr, abs\/2211.05102 (2022)","author":"Pope","year":"2022"},{"key":"ref8","article-title":"Efficient streaming language models with attention sinks","author":"Xiao","year":"2023"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.naacl-long.222"},{"key":"ref10","article-title":"H2o: Heavy-hitter oracle for efficient generative inference of large language models","volume":"36","author":"Zhang","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1043"},{"key":"ref12","article-title":"Scissorhands: Exploiting the persistence of importance hypothesis for llm kv cache compression at test time","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref13","article-title":"Compressive transformers for long-range sequence modelling","author":"Rae","year":"2019"},{"key":"ref14","article-title":"Cam: Cache merging for memory-efficient llms inference","volume-title":"Forty-first International Conference on Machine Learning","author":"Zhang"},{"key":"ref15","article-title":"Model tells you where to merge: Adaptive kv cache merging for llms on long-context tasks","author":"Wang","year":"2024"},{"key":"ref16","article-title":"Dynamic memory compression: Retrofitting llms for accelerated inference","volume-title":"Forty-first International Conference on Machine Learning","author":"Nawrot"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.295"},{"key":"ref18","article-title":"Proof-pile","author":"Azerbayev","year":"2022"},{"key":"ref19","article-title":"The pile: An 800gb dataset of diverse text for language modeling","author":"Gao","year":"2020"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Hyderabad, India","start":{"date-parts":[[2025,4,6]]},"end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10889583.pdf?arnumber=10889583","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T05:25:27Z","timestamp":1774416327000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10889583\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":19,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10889583","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}