{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T06:45:06Z","timestamp":1785912306136,"version":"3.56.0"},"reference-count":64,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Korea government","award":["2710086165"],"award-info":[{"award-number":["2710086165"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6]]},"DOI":"10.1109\/isca66397.2026.00079","type":"proceedings-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T19:16:04Z","timestamp":1785870964000},"page":"1001-1015","source":"Crossref","is-referenced-by-count":0,"title":["SingularBit: Exploiting Synergy of Singular Value Decomposition and Low-Bit Quantization for Weight and KV Compression in LLM Inference"],"prefix":"10.1109","author":[{"given":"Seongyon","family":"Hong","sequence":"first","affiliation":[{"name":"KAIST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hyundeok","family":"Kong","sequence":"additional","affiliation":[{"name":"KAIST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jungwan","family":"Lee","sequence":"additional","affiliation":[{"name":"KAIST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sangjin","family":"Kim","sequence":"additional","affiliation":[{"name":"GIST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hoi-Jun","family":"Yoo","sequence":"additional","affiliation":[{"name":"KAIST"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"OpenAI o3 and o4-mini System Card","year":"2025","journal-title":"OpenAI, Tech. Rep."},{"key":"ref2","article-title":"Deepseek-r1: Incentivizing reasoning capability in 1lms via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Claude Opus 4 & Claude Sonnet 4 System Card","volume-title":"Anthropic, Tech. Rep.","year":"2025"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.507"},{"key":"ref6","article-title":"Let\u2019s verify step by step","volume-title":"The twelfth international conference on learning representations","author":"Lightman","year":"2023"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2019"},{"key":"ref8","article-title":"GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers","volume-title":"International Conference on Learning Representations","author":"Frantar","year":"2023"},{"key":"ref9","first-page":"87","article-title":"Awq: Activation-aware weight quantization for on-device llm compression and acceleration","volume-title":"Proceedings of machine learning and systems","volume":"6","author":"Lin","year":"2024"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1506"},{"key":"ref11","article-title":"Kivi: a tuning-free asymmetric 2bit quantization for kv cache","volume-title":"Proceedings of the 41st International Conference on Machine Learning","author":"Liu","year":"2024"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2198"},{"key":"ref13","article-title":"Language model compression with weighted low-rank factorization","author":"Hsu","year":"2022","journal-title":"arXiv preprint"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1904"},{"key":"ref15","article-title":"Efficient streaming language models with attention sinks","author":"Xiao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0040"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3715996"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731092"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731008"},{"key":"ref20","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref21","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref22","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024","journal-title":"arXiv preprint"},{"key":"ref23","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2016","journal-title":"arXiv preprint"},{"issue":"140","key":"ref24","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"Journal of machine learning research"},{"key":"ref25","article-title":"OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models","volume-title":"International Conference on Learning Representations","author":"Shao","year":"2024"},{"key":"ref26","article-title":"Guidedquant: Large language model quantization via exploiting end loss guidance","volume-title":"International Conference on Machine Learning","author":"Kim","year":"2025"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00266"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.229"},{"key":"ref29","article-title":"Training Verifiers to Solve Math Word Problems","author":"Cobbe","year":"2021","journal-title":"arXiv preprint"},{"key":"ref30","article-title":"Gear: An efficient kv cache compression recipe for nearlossless generative inference of llm","author":"Kang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref31","article-title":"Palu: Compressing kv-cache with low-rank projection","author":"Chang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref32","article-title":"Recalkv: Low-rank kv cache compression via head reordering and offline calibration","author":"Yan","year":"2025","journal-title":"arXiv preprint"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2181"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.172"},{"key":"ref35","first-page":"38087","article-title":"SmoothQuant: Accurate and efficient post-training quantization for large language models","volume-title":"Proceedings of the 40th International Conference on Machine Learning","author":"Xiao","year":"2023"},{"key":"ref36","article-title":"Bitnet: Scaling 1-bit transformers for large language models","author":"Wang","year":"2023","journal-title":"arXiv preprint"},{"key":"ref37","article-title":"The era of 1-bit 11ms: All large language models are in 1.58 bits","author":"Ma","year":"2024","journal-title":"arXiv preprint"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2122"},{"key":"ref39","article-title":"Asvd: Activation-aware singular value decomposition for compressing large language models","author":"Yuan","year":"2025","journal-title":"arXiv preprint"},{"key":"ref40","article-title":"Svd-llm: Truncation-aware singular value decomposition for large language model compression","volume-title":"International Conference on Learning Representations","author":"Wang","year":"2025"},{"key":"ref41","article-title":"Slicegpt: Compress large language models by deleting rows and columns","author":"Ashkboos","year":"2024","journal-title":"arXiv preprint"},{"key":"ref42","article-title":"Matryoshkakv: Adaptive kv compression via trainable orthogonal projection","author":"Lin","year":"2024","journal-title":"arXiv preprint"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731024"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731073"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO61859.2024.00108"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651352"},{"key":"ref47","article-title":"Arb-llm: Alternating refined binarizations for large language models","volume-title":"International Conference on Learning Representations (ICLR)","author":"Li","year":"2025"},{"key":"ref48","article-title":"Are sixteen heads really better than one?","volume-title":"Advances in Neural Information Processing Systems","volume":"32","author":"Michel","year":"2019"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2018.00091"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2018.00069"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731057"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC49661.2025.10904761"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS56072.2025.11043703"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"ref55","article-title":"Think you have solved question answering? try arc, the ai2 reasoning challenge","author":"Clark","year":"2018","journal-title":"arXiv preprint"},{"key":"ref56","first-page":"2924","article-title":"BoolQ: Exploring the surprising difficulty of natural yes\/no questions","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","volume":"1","author":"Clark"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1472"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6399"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2702"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589038"},{"key":"ref62","article-title":"Spqr: A sparsequantized representation for near-lossless llm weight compression","volume-title":"International Conference on Learning Representations (ICLR)","author":"Dettmers","year":"2024"},{"key":"ref63","first-page":"17162","article-title":"Squeezellm: Dense-and-sparse quantization","volume-title":"International Conference on Machine Learning (ICML)","author":"Kim","year":"2024"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731019"}],"event":{"name":"2026 ACM\/IEEE 53rd Annual International Symposium on Computer Architecture (ISCA)","location":"Raleigh, NC, USA","start":{"date-parts":[[2026,6,27]]},"end":{"date-parts":[[2026,7,1]]}},"container-title":["2026 ACM\/IEEE 53rd Annual International Symposium on Computer Architecture (ISCA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11617393\/11617394\/11617476.pdf?arnumber=11617476","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T05:51:29Z","timestamp":1785909089000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11617476\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/isca66397.2026.00079","relation":{},"subject":[],"published":{"date-parts":[[2026,6]]}}}