{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T15:42:13Z","timestamp":1782834133196,"version":"3.54.5"},"publisher-location":"New York, NY, USA","reference-count":7,"publisher":"ACM","funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62572433"],"award-info":[{"award-number":["62572433"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,11,3]]},"DOI":"10.1145\/3680207.3765696","type":"proceedings-article","created":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T13:19:18Z","timestamp":1763731158000},"page":"1389-1391","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Poster: MoE\n                    <sup>2<\/sup>\n                    : Optimizing Collaborative Inference for Edge Large Language Models"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-2454-6577","authenticated-orcid":false,"given":"Lyudong","family":"Jin","sequence":"first","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-0111-2761","authenticated-orcid":false,"given":"Yanning","family":"Zhang","sequence":"additional","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6291-8132","authenticated-orcid":false,"given":"Yanhan","family":"Li","sequence":"additional","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-5221-5043","authenticated-orcid":false,"given":"Shurong","family":"Wang","sequence":"additional","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0256-2416","authenticated-orcid":false,"given":"Howard H.","family":"Yang","sequence":"additional","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3230-6392","authenticated-orcid":false,"given":"Jian","family":"Wu","sequence":"additional","affiliation":[{"name":"Zhejiang University, Hangzhou, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4893-6946","authenticated-orcid":false,"given":"Meng","family":"Zhang","sequence":"additional","affiliation":[{"name":"ZJU-UIUC Institute, Zhejiang University, Haining, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,11,21]]},"reference":[{"key":"e_1_3_2_1_1_1","first-page":"1","article-title":"Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity","volume":"23","author":"Fedus William","year":"2022","unstructured":"William Fedus, Barret Zoph, and Noam Shazeer. 2022. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. Journal of Machine Learning Research 23, 120 (2022), 1\u201339.","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_3_2_1_2_1","volume-title":"Measuring massive multitask language understanding. arXiv preprint arXiv:2009.03300","author":"Hendrycks Dan","year":"2020","unstructured":"Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2020. Measuring massive multitask language understanding. arXiv preprint arXiv:2009.03300 (2020)."},{"key":"e_1_3_2_1_3_1","volume-title":"Adaptive mixtures of local experts. Neural computation 3, 1","author":"Jacobs Robert A","year":"1991","unstructured":"Robert A Jacobs, Michael I Jordan, Steven J Nowlan, and Geoffrey E Hinton. 1991. Adaptive mixtures of local experts. Neural computation 3, 1 (1991), 79\u201387."},{"key":"e_1_3_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2501.09410"},{"key":"e_1_3_2_1_5_1","volume-title":"Hierarchical mixtures of experts and the EM algorithm. Neural computation 6, 2","author":"Jordan Michael I","year":"1994","unstructured":"Michael I Jordan and Robert A Jacobs. 1994. Hierarchical mixtures of experts and the EM algorithm. Neural computation 6, 2 (1994), 181\u2013214."},{"key":"e_1_3_2_1_6_1","unstructured":"Michel Minoux and Hoang Tuy. 2002. Discrete monotonic global optimization. Technical Report. Citeseer."},{"key":"e_1_3_2_1_7_1","volume-title":"Commonsenseqa: A question answering challenge targeting commonsense knowledge. arXiv preprint arXiv:1811.00937","author":"Talmor Alon","year":"2018","unstructured":"Alon Talmor, Jonathan Herzig, Nicholas Lourie, and Jonathan Berant. 2018. Commonsenseqa: A question answering challenge targeting commonsense knowledge. arXiv preprint arXiv:1811.00937 (2018)."}],"event":{"name":"ACM MOBICOM '25: 31st Annual International Conference on Mobile Computing and Networking","location":"Kerry Hotel, Hong Kong Hong Kong China","acronym":"ACM MOBICOM '25","sponsor":["SIGMOBILE ACM Special Interest Group on Mobility of Systems, Users, Data and Computing"]},"container-title":["Proceedings of the 31st Annual International Conference on Mobile Computing and Networking"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3680207.3765696","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T13:22:07Z","timestamp":1763731327000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3680207.3765696"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,3]]},"references-count":7,"alternative-id":["10.1145\/3680207.3765696","10.1145\/3680207"],"URL":"https:\/\/doi.org\/10.1145\/3680207.3765696","relation":{},"subject":[],"published":{"date-parts":[[2025,11,3]]},"assertion":[{"value":"2025-11-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}