{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T07:11:24Z","timestamp":1761894684832,"version":"build-2065373602"},"reference-count":28,"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,6,30]]},"DOI":"10.1109\/icme59968.2025.11210095","type":"proceedings-article","created":{"date-parts":[[2025,10,30]],"date-time":"2025-10-30T17:57:42Z","timestamp":1761847062000},"page":"1-6","source":"Crossref","is-referenced-by-count":0,"title":["Efficient Shared KVCache Attention Inference for Multimodal Large Language Models"],"prefix":"10.1109","author":[{"given":"Shouxu","family":"Kuang","sequence":"first","affiliation":[{"name":"Institute of Software Chinese Academy of Sciences"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Limin","family":"Cheng","sequence":"additional","affiliation":[{"name":"Institute of Software Chinese Academy of Sciences"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yixin","family":"Chen","sequence":"additional","affiliation":[{"name":"Institute of Software Chinese Academy of Sciences"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hang","family":"Qin","sequence":"additional","affiliation":[{"name":"Institute of Software Chinese Academy of Sciences"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ling","family":"Li","sequence":"additional","affiliation":[{"name":"Institute of Software Chinese Academy of Sciences"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Generating images with multimodal language models","volume":"36","author":"Koh","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"article-title":"Text-to-image diffusion models in generative ai: A survey","year":"2023","author":"Zhang","key":"ref2"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3589334.3645633"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.734"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01310"},{"key":"ref6","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":"ref7","first-page":"31094","article-title":"Flexgen: High-throughput generative inference of large language models with a single gpu","volume-title":"International Conference on Machine Learning","author":"Sheng"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref9","first-page":"16344","article-title":"Flashattention: Fast and memory-efficient exact attention with io-awareness","volume":"35","author":"Dao","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.235"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01282"},{"article-title":"Effectively compress kv heads for llm","year":"2024","author":"Yu","key":"ref12"},{"article-title":"Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model","year":"2024","author":"Liu","key":"ref13"},{"article-title":"Cascade inference: Memory bandwidth efficient shared prefix batch decoding","year":"2024","author":"Ye","key":"ref14"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.298"},{"article-title":"Qwen-vl: A frontier large vision-language model with versatile abilities","year":"2023","author":"Bai","key":"ref16"},{"article-title":"Llama 3.2-11B-Vision: A multimodal model by Meta AI","year":"2023","author":"AI","key":"ref17"},{"article-title":"The falcon series of open language models","year":"2023","author":"Almazrouei","key":"ref18"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.696"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2025.3532688"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3503221.3508418"},{"article-title":"Kivi: A tuning-free asymmetric 2bit quantization for kv cache","year":"2024","author":"Liu","key":"ref22"},{"article-title":"Sequence can secretly tell you what to discard","year":"2024","author":"Dai","key":"ref23"},{"key":"ref24","article-title":"Llava-med: Training a large language-and-vision assistant for biomedicine in one day","volume":"36","author":"Li","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72643-9_22"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"article-title":"Milebench: Benchmarking mllms in long context","year":"2024","author":"Song","key":"ref27"},{"article-title":"Qwen2. 5-coder technical report","year":"2024","author":"Hui","key":"ref28"}],"event":{"name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","start":{"date-parts":[[2025,6,30]]},"location":"Nantes, France","end":{"date-parts":[[2025,7,4]]}},"container-title":["2025 IEEE International Conference on Multimedia and Expo (ICME)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11208895\/11208897\/11210095.pdf?arnumber=11210095","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T05:33:26Z","timestamp":1761888806000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11210095\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,30]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1109\/icme59968.2025.11210095","relation":{},"subject":[],"published":{"date-parts":[[2025,6,30]]}}}