{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,9,7]],"date-time":"2024-09-07T20:50:54Z","timestamp":1725742254970},"reference-count":26,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,1,28]],"date-time":"2024-01-28T00:00:00Z","timestamp":1706400000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,1,28]],"date-time":"2024-01-28T00:00:00Z","timestamp":1706400000000},"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":[[2024,1,28]]},"DOI":"10.1109\/iceic61013.2024.10457111","type":"proceedings-article","created":{"date-parts":[[2024,3,19]],"date-time":"2024-03-19T18:06:51Z","timestamp":1710871611000},"page":"1-4","source":"Crossref","is-referenced-by-count":0,"title":["Searching Optimal Floating-Point Format for Sub-8-Bit Large Language Model Inference"],"prefix":"10.1109","author":[{"given":"Youngdeok","family":"Hwang","sequence":"first","affiliation":[{"name":"City University of New York"}]},{"given":"Janghwan","family":"Lee","sequence":"additional","affiliation":[{"name":"Hanyang University"}]},{"given":"Jiwoong","family":"Park","sequence":"additional","affiliation":[{"name":"Hanyang University"}]},{"given":"Jieun","family":"Lim","sequence":"additional","affiliation":[{"name":"SAPEON Korea Inc."}]},{"given":"Jungwook","family":"Choi","sequence":"additional","affiliation":[{"name":"Hanyang University"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Opt: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv preprint"},{"key":"ref2","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Bloom: A 176b-parameter open-access multilingual language model","author":"Scao","year":"2022","journal-title":"arXiv preprint"},{"journal-title":"Gpt-neo: Large scale autoregressive language modeling with mesh-tensorflow","year":"2021","author":"Black","key":"ref4"},{"key":"ref5","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2019","journal-title":"NAACL-HLT"},{"key":"ref6","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"NeurIPS"},{"key":"ref7","article-title":"1.1 computings energy problem","author":"Horowitz","year":"2014","journal-title":"ISSCC"},{"journal-title":"Nvidia hopper architecture in-depth","year":"2022","author":"Andersch","key":"ref8"},{"key":"ref9","article-title":"LIm. int8 (): 8-bit matrix multiplication for trans-formers at scale","author":"Dettmers","year":"2022","journal-title":"NeurIPS"},{"key":"ref10","article-title":"Smoothquant: Accurate and efficient post-training quantization for large language models","author":"Xiao","year":"2023","journal-title":"ICML"},{"key":"ref11","article-title":"Omniquant: Omnidirectionally calibrated quantization for large language models","author":"Shao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref12","article-title":"Efficient post-training quantization with fp8 formats","author":"Shen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref13","article-title":"Training and inference of large language models using 8-bit floating point","author":"Perez","year":"2023","journal-title":"arXiv preprint"},{"key":"ref14","article-title":"Exploring the potential of flexible 8-bit format: Design and algorithm","author":"Zhang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref15","article-title":"Fp8 quantization: The power of the exponent","author":"Kuzmin","year":"2022","journal-title":"NeurIPS"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.39"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096798"},{"key":"ref18","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2017","journal-title":"ICLR"},{"key":"ref19","article-title":"Attention is all you need","author":"Vaswani","year":"2017","journal-title":"NeurIPS"},{"key":"ref20","article-title":"Mixed precision training","author":"Micikevicius","year":"2018","journal-title":"ICLR"},{"key":"ref21","article-title":"Hybrid 8-bit floating point (hfp8) training and inference for deep neural networks","author":"Sun","year":"2019","journal-title":"NeurIPS"},{"key":"ref22","article-title":"Ultra-low precision 4-bit training of deep neural networks","author":"Sun","year":"2020","journal-title":"NeurIPS"},{"key":"ref23","article-title":"Introducing mpt-7b: A new standard for open-source","author":"Team","year":"2023","journal-title":"commercially usable llms"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3474381"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"ref26","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","author":"Raffel","year":"2019","journal-title":"arXiv e-prints"}],"event":{"name":"2024 International Conference on Electronics, Information, and Communication (ICEIC)","start":{"date-parts":[[2024,1,28]]},"location":"Taipei, Taiwan","end":{"date-parts":[[2024,1,31]]}},"container-title":["2024 International Conference on Electronics, Information, and Communication (ICEIC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10457047\/10457087\/10457111.pdf?arnumber=10457111","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,26]],"date-time":"2024-03-26T14:11:45Z","timestamp":1711462305000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10457111\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,1,28]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1109\/iceic61013.2024.10457111","relation":{},"subject":[],"published":{"date-parts":[[2024,1,28]]}}}