{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T05:04:02Z","timestamp":1780031042838,"version":"3.53.1"},"reference-count":50,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,4,8]],"date-time":"2026-04-08T00:00:00Z","timestamp":1775606400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,4,8]],"date-time":"2026-04-08T00:00:00Z","timestamp":1775606400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,4,8]]},"DOI":"10.1109\/isqed69900.2026.11534707","type":"proceedings-article","created":{"date-parts":[[2026,5,28]],"date-time":"2026-05-28T22:26:58Z","timestamp":1780007218000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["CELLM: Confidential and Efficient Lightweight LLM Inference in TEEs"],"prefix":"10.1109","author":[{"given":"Hui","family":"Feng","sequence":"first","affiliation":[{"name":"University of California,Electrical Engineering and Computer Science,Merced,CA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ben","family":"Dong","sequence":"additional","affiliation":[{"name":"University of California,Electrical Engineering and Computer Science,Merced,CA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qian","family":"Wang","sequence":"additional","affiliation":[{"name":"University of California,Electrical Engineering and Computer Science,Merced,CA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.hcc.2024.100211"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i22.34537"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.naacl-main.13"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.5555\/3241094.3241142"},{"key":"ref5","first-page":"723","article-title":"SOTER: Guarding black-box inference for general neural networks at the edge","volume-title":"2022 USENIX Annual Technical Conference (USENIX ATC 22)","author":"Shen"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/SP46215.2023.10179382"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.543"},{"key":"ref8","first-page":"1939","article-title":"Scalable multi-party computation protocols for machine learning in the honest-majority setting","volume-title":"Proc. 33rd USENIX Secur. Symp.(USENIX Secur.)","author":"Liu"},{"key":"ref9","article-title":"Intel\u00ae trust domain extensions","year":"2022"},{"key":"ref10","article-title":"Overview on signing and whitelisting for intel\u00ae software guard extensions (intel\u00ae sgx) enclaves","year":"2019"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3652597"},{"key":"ref12","article-title":"Amd memory encryption: Sev, sme, and sev-es","year":"2018"},{"key":"ref13","article-title":"Building a secure system using trustzone technology","year":"2020"},{"key":"ref14","first-page":"465","article-title":"Design and verification of the arm confidential compute architecture","volume-title":"16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)","author":"Li"},{"key":"ref15","article-title":"Evaluating the performance of the deepseek model in confidential computing environment","volume-title":"arXiv preprint arXiv:2502.11347","author":"Dong"},{"key":"ref16","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Advances in Neural Information Processing Systems (NeurIPS)"},{"key":"ref17","article-title":"Introducing gemini: Our most capable ai model","author":"DeepMind","year":"2023"},{"key":"ref18","article-title":"Mobillama: Towards accurate and lightweight fully transparent gpt","author":"Thawakar","year":"2024","journal-title":"arXiv preprint arXiv:2402.16840"},{"key":"ref19","article-title":"Gemma 2: Improving open language models at a practical size","author":"Team","year":"2024","journal-title":"arXiv preprint arXiv:2408.00118"},{"key":"ref20","article-title":"Deepseek-r1: Incentivizing reasoning capability in 11 ms via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv preprint arXiv:2501.12948"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1319"},{"key":"ref22","article-title":"Deepseek-v3 technical report","volume-title":"arXiv preprint arXiv:2412.19437","author":"Liu","year":"2024"},{"issue":"2","key":"ref23","article-title":"A survey of large language models","volume":"1","author":"Zhao","year":"2023","journal-title":"arXiv preprint arXiv:2303.18223"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3300061.3345447"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3386901.3388946"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/SP54263.2024.00052"},{"key":"ref27","article-title":"Realsafe-r1: Safety-aligned deepseek-r1 without compromising reasoning capability","author":"Zhang","year":"2025","journal-title":"arXiv preprint arXiv:2504.10081"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0040"},{"key":"ref29","first-page":"38087","article-title":"Smoothquant: Accurate and efficient post-training quantization for large language models","volume-title":"International conference on machine learning. PMLR","author":"Xiao"},{"key":"ref30","article-title":"Generating long sequences with sparse transformers","author":"Child","year":"2019","journal-title":"arXiv preprint arXiv:1904.10509"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3411508.3421376"},{"key":"ref32","article-title":"Confidential inference via ternary model partitioning","volume-title":"arXiv preprint arXiv:1807.00969","author":"Gu"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3423211.3425687"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3419111.3421282"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3707453"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD57390.2023.10323746"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/DAC63849.2025.11132467"},{"key":"ref38","article-title":"Ensembler: Protect collaborative inference privacy from model inversion attack via selective ensemble","author":"Liu","year":"2024","journal-title":"arXiv e-prints, pp. arXiv-2401"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731047"},{"key":"ref40","article-title":"Deepseek-r1-distill-qwen-1.5b","year":"2025"},{"key":"ref41","article-title":"Llama-3.2-1b","year":"2024","journal-title":"hugging Face model repository."},{"key":"ref42","article-title":"Llama-3.2-3b","year":"2024","journal-title":"hugging Face model repository"},{"key":"ref43","article-title":"Gemma-2-2b","year":"2024","journal-title":"hugging Face model repository"},{"key":"ref44","article-title":"Deepseek-r1-distill-qwen-7b","year":"2025","journal-title":"hugging Face model repository."},{"key":"ref45","article-title":"Deepseek-r1-distill-qwen-14b","year":"2025","journal-title":"hugging Face model repository"},{"key":"ref46","article-title":"Amc23 dataset","author":"He","year":"2024","journal-title":"hugging Face dataset repository."},{"key":"ref47","first-page":"158","article-title":"Program induction by rationale generation: Learning to solve and explain algebraic word problems","volume-title":"Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics","volume":"1","author":"Ling"},{"key":"ref48","article-title":"Aime 2024 dataset","author":"Jia","year":"2024","journal-title":"hugging Face dataset repository."},{"key":"ref49","article-title":"Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning","year":"2025"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC66894.2025.00017"}],"event":{"name":"2026 27th International Symposium on Quality Electronic Design (ISQED)","location":"San Francisco, CA, USA","start":{"date-parts":[[2026,4,8]]},"end":{"date-parts":[[2026,4,10]]}},"container-title":["2026 27th International Symposium on Quality Electronic Design (ISQED)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11534651\/11534652\/11534707.pdf?arnumber=11534707","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T04:25:19Z","timestamp":1780028719000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11534707\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,8]]},"references-count":50,"URL":"https:\/\/doi.org\/10.1109\/isqed69900.2026.11534707","relation":{},"subject":[],"published":{"date-parts":[[2026,4,8]]}}}