{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,15]],"date-time":"2025-11-15T07:28:38Z","timestamp":1763191718135,"version":"3.45.0"},"reference-count":32,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,6,30]],"date-time":"2025-06-30T00:00:00Z","timestamp":1751241600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,30]],"date-time":"2025-06-30T00:00:00Z","timestamp":1751241600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004735","name":"Natural Science Foundation of Hunan Province","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004735","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,6,30]]},"DOI":"10.1109\/ijcnn64981.2025.11228214","type":"proceedings-article","created":{"date-parts":[[2025,11,14]],"date-time":"2025-11-14T18:46:15Z","timestamp":1763145975000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["Initial-Key Cache: An Efficient KV Cache Strategy Focusing on Initial and Key Tokens for LLMs"],"prefix":"10.1109","author":[{"given":"Zhongyi","family":"Tang","sequence":"first","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Zejiang","family":"He","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Junzhong","family":"Shen","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Yiyue","family":"Hu","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Luchen","family":"Zhou","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Yongzhang","family":"Nie","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]},{"given":"Yongwen","family":"Wang","sequence":"additional","affiliation":[{"name":"National University of Defense Technology,College of Computer Science and Technology,China"}]}],"member":"263","reference":[{"article-title":"Large language models: A survey","year":"2024","author":"Minaee","key":"ref1"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3357384.3357895"},{"article-title":"Controllable text generation for large language models: A survey","year":"2024","author":"Liang","key":"ref3"},{"key":"ref4","first-page":"41 092","article-title":"Prompting large language model for machine translation: A case study","volume-title":"International Conference on Machine Learning","author":"Zhang"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.307"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.172"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"article-title":"Efficient streaming language models with attention sinks","year":"2023","author":"Xiao","key":"ref8"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N19-4009"},{"article-title":"Llama: Open and efficient foundation language models","year":"2023","author":"Touvron","key":"ref10"},{"article-title":"Opt: Open pre-trained transformer language models","year":"2022","author":"Zhang","key":"ref11"},{"key":"ref12","first-page":"2397","article-title":"Pythia: A suite for analyzing large language models across training and scaling","volume-title":"International Conference on Machine Learning","author":"Biderman"},{"article-title":"Qwen technical report","year":"2023","author":"Bai","key":"ref13"},{"article-title":"Longformer: The long-document transformer","year":"2020","author":"Beltagy","key":"ref14"},{"article-title":"Longnet: Scaling transformers to 1,000,000,000 tokens","year":"2023","author":"Ding","key":"ref15"},{"article-title":"Generating long sequences with sparse transformers","year":"2019","author":"Child","key":"ref16"},{"key":"ref17","first-page":"606","article-title":"Efficiently scaling transformer inference","volume-title":"Proceedings of Machine Learning and Systems","volume":"5","author":"Pope"},{"key":"ref18","first-page":"155","article-title":"{InfiniGen} : Efficient generative inference of large language models with dynamic {KV} cache management","volume-title":"18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)","author":"Lee"},{"article-title":"Zero-delay qkv compression for mitigating kv cache and network bottlenecks in llm inference","year":"2024","author":"Zhang","key":"ref19"},{"article-title":"Gear: An efficient kv cache compression recipefor near-lossless generative inference of llm","year":"2024","author":"Kang","key":"ref20"},{"article-title":"Kivi: A tuning-free asymmetric 2bit quantization for kv cache","year":"2024","author":"Liu","key":"ref21"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672274"},{"key":"ref23","first-page":"34 661","article-title":"H2o: Heavy-hitter oracle for efficient generative inference of large language models","volume-title":"Advances in Neural Information Processing Systems","volume":"36","author":"Zhang"},{"key":"ref24","first-page":"114","article-title":"Keyformer: Kv cache reduction through key tokens selection for efficient generative inference","volume-title":"Proceedings of Machine Learning and Systems","volume":"6","author":"Adnan"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00077"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.1079"},{"key":"ref27","first-page":"8","article-title":"A framework for few-shot language model evaluation","volume":"10","author":"Gao","year":"2021","journal-title":"Version v0. 0.1"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21236\/ADA273556"},{"article-title":"Pointer sentinel mixture models","year":"2016","author":"Merity","key":"ref29"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1260"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3581783.3611826"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"}],"event":{"name":"2025 International Joint Conference on Neural Networks (IJCNN)","start":{"date-parts":[[2025,6,30]]},"location":"Rome, Italy","end":{"date-parts":[[2025,7,5]]}},"container-title":["2025 International Joint Conference on Neural Networks (IJCNN)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11227166\/11227148\/11228214.pdf?arnumber=11228214","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,15]],"date-time":"2025-11-15T07:26:55Z","timestamp":1763191615000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11228214\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,30]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/ijcnn64981.2025.11228214","relation":{},"subject":[],"published":{"date-parts":[[2025,6,30]]}}}