{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T14:17:16Z","timestamp":1781360236460,"version":"3.54.1"},"publisher-location":"New York, NY, USA","reference-count":29,"publisher":"ACM","license":[{"start":{"date-parts":[[2024,10,27]],"date-time":"2024-10-27T00:00:00Z","timestamp":1729987200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2024,10,27]]},"DOI":"10.1145\/3676536.3676776","type":"proceedings-article","created":{"date-parts":[[2025,4,9]],"date-time":"2025-04-09T12:53:56Z","timestamp":1744203236000},"page":"1-9","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Towards Floating Point-Based Attention-Free LLM: Hybrid PIM with Non-Uniform Data Format and Reduced Multiplications"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4162-6360","authenticated-orcid":false,"given":"Lidong","family":"Guo","sequence":"first","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9259-7180","authenticated-orcid":false,"given":"Zhenhua","family":"Zhu","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-8943-1577","authenticated-orcid":false,"given":"Tengxuan","family":"Liu","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"},{"name":"Infinigence-AI, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2209-8312","authenticated-orcid":false,"given":"Xuefei","family":"Ning","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-4755-6881","authenticated-orcid":false,"given":"Shiyao","family":"Li","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"},{"name":"Infinigence-AI, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0849-3252","authenticated-orcid":false,"given":"Guohao","family":"Dai","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"},{"name":"Infinigence-AI, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2421-353X","authenticated-orcid":false,"given":"Huazhong","family":"Yang","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3185-0457","authenticated-orcid":false,"given":"Wangyang","family":"Fu","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6108-5157","authenticated-orcid":false,"given":"Yu","family":"Wang","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,4,9]]},"reference":[{"key":"e_1_3_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2023.3272658"},{"key":"e_1_3_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"e_1_3_2_1_3_1","unstructured":"Tom Brown et al. 2020. Language models are few-shot learners. Advances in neural information processing systems 33 1877--1901."},{"key":"e_1_3_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001140"},{"key":"e_1_3_2_1_5_1","volume-title":"2021 IEEE International Solid-State Circuits Conference (ISSCC).","volume":"64","author":"Yu-Der","unstructured":"Yu-Der Chih et al. 2021. 16.4 an 89tops\/w and 16.3 tops\/mm 2 all-digital sram-based full-precision compute-in memory macro in 22nm for machine-learning edge applications. In 2021 IEEE International Solid-State Circuits Conference (ISSCC). Vol. 64. IEEE, 252--254."},{"key":"e_1_3_2_1_6_1","unstructured":"Peter Clark Isaac Cowhey Oren Etzioni Tushar Khot Ashish Sabharwal Carissa Schoenick and Oyvind Tafjord. 2018. Think you have solved question answering? try arc the ai2 reasoning challenge. arXiv preprint arXiv:1803.05457."},{"key":"e_1_3_2_1_7_1","unstructured":"Tim Dettmers Artidoro Pagnoni Ari Holtzman and Luke Zettlemoyer. 2024. Qlora: efficient finetuning of quantized llms. Advances in Neural Information Processing Systems 36."},{"key":"e_1_3_2_1_8_1","unstructured":"Daniel Y Fu Tri Dao Khaled K Saab Armin W Thomas Atri Rudra and Christopher R\u00e9. 2022. Hungry hungry hippos: towards language modeling with state space models. arXiv preprint arXiv:2212.14052."},{"key":"e_1_3_2_1_9_1","unstructured":"Albert Gu and Tri Dao. 2023. Mamba: linear-time sequence modeling with selective state spaces. arXiv preprint arXiv:2312.00752."},{"key":"e_1_3_2_1_10_1","unstructured":"Albert Gu Karan Goel and Christopher R\u00e9. 2021. Efficiently modeling long sequences with structured state spaces. arXiv preprint arXiv:2111.00396."},{"key":"e_1_3_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2021.3064189"},{"key":"e_1_3_2_1_12_1","unstructured":"Shih-yang Liu Zechun Liu Xijie Huang Pingcheng Dong and Kwang-Ting Cheng. 2023. Llm-fp4: 4-bit floating-point quantized transformers. arXiv preprint arXiv:2310.16836."},{"key":"e_1_3_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2007.33"},{"key":"e_1_3_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3453688.3461509"},{"key":"e_1_3_2_1_15_1","volume-title":"Raffaella Bernardi, Sandro Pezzelle, Marco Baroni, Gemma Boleda, and Raquel Fern\u00e1ndez.","author":"Paperno Denis","year":"2016","unstructured":"Denis Paperno, Germ\u00e1n Kruszewski, Angeliki Lazaridou, Quan Ngoc Pham, Raffaella Bernardi, Sandro Pezzelle, Marco Baroni, Gemma Boleda, and Raquel Fern\u00e1ndez. 2016. The lambada dataset: word prediction requiring a broad discourse context. arXiv preprint arXiv:1606.06031."},{"key":"e_1_3_2_1_16_1","unstructured":"Bo Peng et al. 2023. Rwkv: reinventing rnns for the transformer era. arXiv preprint arXiv:2305.13048."},{"key":"e_1_3_2_1_17_1","volume-title":"International Conference on Machine Learning. PMLR, 28043--28078","author":"Poli Michael","year":"2023","unstructured":"Michael Poli, Stefano Massaroli, Eric Nguyen, Daniel Y Fu, Tri Dao, Stephen Baccus, Yoshua Bengio, Stefano Ermon, and Christopher R\u00e9. 2023. Hyena hierarchy: towards larger convolutional language models. In International Conference on Machine Learning. PMLR, 28043--28078."},{"key":"e_1_3_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3474381"},{"key":"e_1_3_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001139"},{"key":"e_1_3_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/LSSC.2023.3252607"},{"key":"e_1_3_2_1_21_1","unstructured":"Yutao Sun Li Dong Shaohan Huang Shuming Ma Yuqing Xia Jilong Xue Jianyong Wang and Furu Wei. 2023. Retentive network: a successor to transformer for large language models. arXiv preprint arXiv:2307.08621."},{"key":"e_1_3_2_1_22_1","first-page":"243","article-title":"Redcim: reconfigurable digital computing-in-memory processor with unified fp\/int pipeline for cloud ai acceleration","volume":"58","author":"Fengbin Tu","year":"2022","unstructured":"Fengbin Tu et al. 2022. Redcim: reconfigurable digital computing-in-memory processor with unified fp\/int pipeline for cloud ai acceleration. IEEE Journal of Solid-State Circuits, 58, 1, 243--255.","journal-title":"IEEE Journal of Solid-State Circuits"},{"key":"e_1_3_2_1_23_1","unstructured":"Ashish Vaswani Noam Shazeer Niki Parmar Jakob Uszkoreit Llion Jones Aidan N Gomez \u0141ukasz Kaiser and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems 30."},{"key":"e_1_3_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD51958.2021.9643573"},{"key":"e_1_3_2_1_25_1","volume-title":"2022 IEEE International Solid-State Circuits Conference (ISSCC).","volume":"65","author":"Bonan","unstructured":"Bonan Yan et al. 2022. A 1.041-mb\/mm 2 27.38-tops\/wsigned-int8 dynamic-logic-based adc-less sram compute-in-memory macro in 28nm with reconfigurable bitwise operation for ai and embedded applications. In 2022 IEEE International Solid-State Circuits Conference (ISSCC). Vol. 65. IEEE, 188--190."},{"key":"e_1_3_2_1_26_1","doi-asserted-by":"crossref","unstructured":"Rowan Zellers Ari Holtzman Yonatan Bisk Ali Farhadi and Yejin Choi. 2019. Hellaswag: can a machine really finish your sentence? arXiv preprint arXiv:1905.07830.","DOI":"10.18653\/v1\/P19-1472"},{"key":"e_1_3_2_1_27_1","unstructured":"Shuangfei Zhai Walter Talbott Nitish Srivastava Chen Huang Hanlin Goh Ruixiang Zhang and Josh Susskind. 2021. An attention free transformer. arXiv preprint arXiv:2105.14103."},{"key":"e_1_3_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3316781.3317739"},{"key":"e_1_3_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2023.3251696"}],"event":{"name":"ICCAD '24: 43rd IEEE\/ACM International Conference on Computer-Aided Design","location":"Newark Liberty International Airport Marriott New York NY USA","acronym":"ICCAD '24","sponsor":["SIGDA ACM Special Interest Group on Design Automation","IEEE CAS","IEEE CEDA","IEEE EDS"]},"container-title":["Proceedings of the 43rd IEEE\/ACM International Conference on Computer-Aided Design"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3676536.3676776","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3676536.3676776","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T23:43:58Z","timestamp":1750290238000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3676536.3676776"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,27]]},"references-count":29,"alternative-id":["10.1145\/3676536.3676776","10.1145\/3676536"],"URL":"https:\/\/doi.org\/10.1145\/3676536.3676776","relation":{},"subject":[],"published":{"date-parts":[[2024,10,27]]},"assertion":[{"value":"2025-04-09","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}