{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T05:24:59Z","timestamp":1747373099337,"version":"3.28.0"},"reference-count":21,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,8,11]],"date-time":"2024-08-11T00:00:00Z","timestamp":1723334400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,8,11]],"date-time":"2024-08-11T00:00:00Z","timestamp":1723334400000},"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":[[2024,8,11]]},"DOI":"10.1109\/mwscas60917.2024.10658840","type":"proceedings-article","created":{"date-parts":[[2024,9,16]],"date-time":"2024-09-16T17:34:29Z","timestamp":1726508069000},"page":"243-247","source":"Crossref","is-referenced-by-count":2,"title":["Hessian-Aware KV Cache Quantization for LLMs"],"prefix":"10.1109","author":[{"given":"Woohong","family":"Byun","sequence":"first","affiliation":[{"name":"School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,Georgia,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jongseok","family":"Woo","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,Georgia,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Saibal","family":"Mukhopadhyay","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, Georgia Institute of Technology,Atlanta,Georgia,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref2","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/MWSCAS57524.2023.10406106"},{"key":"ref4","article-title":"Robust processing-in-memory with multi-bit reram using hessian-driven mixed-precision computation","author":"Dash","year":"2021","journal-title":"IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst."},{"journal-title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","year":"2018","author":"Devlin","key":"ref5"},{"key":"ref6","doi-asserted-by":"crossref","DOI":"10.1145\/3400302.3415679","article-title":"Hessian-Driven DNN Parameter Protection","volume-title":"Proc. IEEE\/ACM ICCAD","author":"Dash","year":"2020"},{"issue":"2018","key":"ref7","first-page":"58","article-title":"Accelerated stochastic power iteration","volume-title":"Proc. 21st Int. Conf. Artif. Intell. and Statist., Mach. Learn. Res.","volume":"84","author":"Xu"},{"journal-title":"pytorch-hessian-eigenthings: efficient PyTorch Hessian eigendecomposition","year":"2018","author":"Golmant","key":"ref8"},{"key":"ref9","article-title":"KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref10","article-title":"KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization","author":"Hooper","year":"2024","journal-title":"arXiv preprint"},{"key":"ref11","article-title":"A framework for few-shot language model evaluation","author":"Gao","year":"2021","journal-title":"arXiv preprint, Version v0. 0.1"},{"key":"ref12","article-title":"Longbench: A bilingual, multitask benchmark for long context understanding","author":"Bai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref13","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"volume-title":"New models and developer products announced at DevDay 2023","year":"2023","key":"ref14"},{"key":"ref15","article-title":"LongLORA: Efficient Fine-Tuning of Long-Context Large Language Models","author":"Chen","year":"2023","journal-title":"arXiv preprint"},{"key":"ref16","article-title":"Efficiently Scaling Transformer Inference","volume-title":"Proc. of Machine Learning and Systems","volume":"5","author":"Pope","year":"2023"},{"key":"ref17","article-title":"Scaling Infrastructure for Multi-Trillion Parameter LLM Training","volume-title":"Proc. ASSYST@ ISCA","author":"Isaev","year":"2023"},{"key":"ref18","article-title":"Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache","author":"Lin","year":"2024","journal-title":"arXiv preprint"},{"key":"ref19","article-title":"Atom: Low-bit quantization for efficient and accurate LLM serving","author":"Zhao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref20","article-title":"FlexGen: High-throughput generative inference of large language models with a single GPU","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Sheng","year":"2023"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3665314.3670806"}],"event":{"name":"2024 IEEE 67th International Midwest Symposium on Circuits and Systems (MWSCAS)","start":{"date-parts":[[2024,8,11]]},"location":"Springfield, MA, USA","end":{"date-parts":[[2024,8,14]]}},"container-title":["2024 IEEE 67th International Midwest Symposium on Circuits and Systems (MWSCAS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10654782\/10654792\/10658840.pdf?arnumber=10658840","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,20]],"date-time":"2024-09-20T05:32:05Z","timestamp":1726810325000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10658840\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,11]]},"references-count":21,"URL":"https:\/\/doi.org\/10.1109\/mwscas60917.2024.10658840","relation":{},"subject":[],"published":{"date-parts":[[2024,8,11]]}}}