{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T21:17:35Z","timestamp":1773263855319,"version":"3.50.1"},"reference-count":43,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,1,19]],"date-time":"2026-01-19T00:00:00Z","timestamp":1768780800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,19]],"date-time":"2026-01-19T00:00:00Z","timestamp":1768780800000},"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":[[2026,1,19]]},"DOI":"10.1109\/asp-dac66049.2026.11420427","type":"proceedings-article","created":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T19:51:15Z","timestamp":1773172275000},"page":"716-722","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Trustworthiness with Mixed Precision: Benchmarks, Opportunities, and Challenges"],"prefix":"10.1109","author":[{"given":"Guanxi","family":"Lu","sequence":"first","affiliation":[{"name":"Imperial College,Department of Electrical and Electronic Engineering,London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao Mark","family":"Chen","sequence":"additional","affiliation":[{"name":"Imperial College= Imperial College,Department of ComputingEngineering,London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhiqiang","family":"Que","sequence":"additional","affiliation":[{"name":"Imperial College= Imperial College,Department of ComputingEngineering,London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wayne","family":"Luk","sequence":"additional","affiliation":[{"name":"Imperial College= Imperial College,Department of ComputingEngineering,London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hongxiang","family":"Fan","sequence":"additional","affiliation":[{"name":"Imperial College= Imperial College,Department of ComputingEngineering,London"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Gpt-4 technical report","volume-title":"arXiv preprint arXiv:2303.08774","author":"Achiam","year":"2023"},{"key":"ref2","article-title":"Qwen3 technical report","volume-title":"arXiv preprint arXiv:2505.09388","author":"Yang","year":"2025"},{"key":"ref3","first-page":"87","article-title":"Awq: Activation-aware weight quantization for on-device 11 m compression and acceleration","volume-title":"Proceedings of machine learning and systems","volume":"6","author":"Lin"},{"key":"ref4","article-title":"Zeroquant: Efficient and affordable post-training quantization for largescale transformers","author":"Yao"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0040"},{"key":"ref6","article-title":"Decodingtrust: A comprehensive assessment of trustworthiness in gpt models","author":"Wang","year":"2023"},{"key":"ref7","article-title":"A white paper on neural network quantization","author":"Nagel","year":"2021","journal-title":"arXiv preprint arXiv:2106.08295"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/fpl.2018.00056"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICFPT47387.2019.00012"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3116302"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1201\/9781003162810-13"},{"key":"ref13","article-title":"Gptq: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022","journal-title":"arXiv preprint arXiv:2210.17323"},{"key":"ref14","first-page":"23901","article-title":"Squeezellm: dense-and-sparse quantization","volume-title":"Proceedings of the 41st International Conference on Machine Learning","author":"Kim"},{"key":"ref15","first-page":"39682","article-title":"Any-precision 1lm: low-cost deployment of multiple, different-sized llms","volume-title":"Proceedings of the 41st International Conference on Machine Learning","author":"Park"},{"key":"ref16","article-title":"Progressive mixed-precision decoding for efficient 11 m inference","volume-title":"The Thirteenth International Conference on Learning Representations.","author":"Chen"},{"key":"ref17","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"},{"key":"ref18","article-title":"Rptq: Reorder-based post-training quantization for large language models","author":"Yuan","year":"2023","journal-title":"arXiv preprint arXiv:2304.01089"},{"key":"ref19","first-page":"32332","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"},{"key":"ref20","article-title":"Wkvquant: Quantizing weight and key\/value cache for large language models gains more","author":"Yue","year":"2024"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01670-7"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247954"},{"key":"ref23","first-page":"325","article-title":"Enabling fast uncertainty estimation: accelerating bayesian transformers via algorithmic and hardware optimizations","volume-title":"Proceedings of the 59th ACM\/IEEE Design Automation Conference","author":"Fan"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.512"},{"key":"ref25","article-title":"Trustllm: Trustworthiness in large language models","author":"Huang"},{"key":"ref26","article-title":"Trustworthy 11 ms: a survey and guideline for evaluating large language models\u2019 alignment","author":"Liu","year":"2023","journal-title":"arXiv preprint arXiv:2308.05374"},{"key":"ref27","article-title":"Trism for agentic ai: A review of trust, risk, and security management in 11 m based agentic multi-agent systems","author":"Raza","year":"2025","journal-title":"arXiv preprint arXiv:2506.04133"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-emnlp.356"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2024\/902"},{"key":"ref30","article-title":"Decoding compressed trust: Scrutinizing the trustworthiness of efficient llms under compression","author":"Hong","year":"2024","journal-title":"arXiv:2403.15447"},{"key":"ref31","article-title":"Smaller= weaker? benchmarking robustness of quantized 11 ms in code generation","author":"Fang","year":"2025","journal-title":"arXiv preprint arXiv:2506.22776"},{"key":"ref32","article-title":"Harmlevelbench: Evaluating harm-level compliance and the impact of quantization on model alignment","author":"Belkhiter","year":"2024","journal-title":"arXiv preprint arXiv:2411.06835"},{"key":"ref33","article-title":"Quantized but deceptive? a multi-dimensional truthfulness evaluation of quantized 11 ms","author":"Fu","year":"2025","journal-title":"arXiv preprint arXiv:2508.19432"},{"key":"ref34","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv:2307.09288"},{"key":"ref35","article-title":"Investigating the impact of quantization methods on the safety and reliability of large language models","author":"Kharinaev","year":"2025","journal-title":"arXiv preprint arXiv:2502.15799"},{"key":"ref36","article-title":"Measuring massive multitask language understanding","author":"Hendrycks","year":"2020","journal-title":"arXiv preprint arXiv:2009.03300"},{"key":"ref37","article-title":"Adversarial glue: A multi-task benchmark for robustness evaluation of language models","volume-title":"Advances in Neural Information Processing Systems","author":"Wang"},{"key":"ref38","author":"Becker","year":"1996","journal-title":"Adult,\u201d UCI Machine Learning Repository"},{"key":"ref39","article-title":"Aligning ai with shared human values","volume-title":"International Conference on Learning Representations","author":"Hendrycks"},{"key":"ref40","article-title":"Scaling llm test-time compute optimally can be more effective than scaling model parameters","author":"Snell","year":"2024","journal-title":"arXiv preprint arXiv:2408.03314"},{"key":"ref41","article-title":"Rethinking optimal verification granularity for compute-efficient testtime scaling","author":"Chen","year":"2025","journal-title":"arXiv preprint arXiv:2505.11730"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.138"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2022.3160948"}],"event":{"name":"2026 31st Asia and South Pacific Design Automation Conference (ASP-DAC)","location":"Lantau, Hong Kong","start":{"date-parts":[[2026,1,19]]},"end":{"date-parts":[[2026,1,22]]}},"container-title":["2026 31st Asia and South Pacific Design Automation Conference (ASP-DAC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11420221\/11420229\/11420427.pdf?arnumber=11420427","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T05:17:32Z","timestamp":1773206252000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11420427\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,19]]},"references-count":43,"URL":"https:\/\/doi.org\/10.1109\/asp-dac66049.2026.11420427","relation":{},"subject":[],"published":{"date-parts":[[2026,1,19]]}}}