{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,6]],"date-time":"2025-11-06T19:00:36Z","timestamp":1762455636227,"version":"build-2065373602"},"reference-count":71,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"MSRA"},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62076184","62473286"],"award-info":[{"award-number":["62076184","62473286"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007219","name":"Natural Science Foundation of Shanghai Municipality","doi-asserted-by":"publisher","award":["22ZR1466700"],"award-info":[{"award-number":["22ZR1466700"]}],"id":[{"id":"10.13039\/100007219","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2025,12]]},"DOI":"10.1109\/tpami.2025.3598206","type":"journal-article","created":{"date-parts":[[2025,8,12]],"date-time":"2025-08-12T18:01:55Z","timestamp":1755021715000},"page":"11596-11613","source":"Crossref","is-referenced-by-count":0,"title":["EA-HAS-Bench and Language-Enhanced Shrinkage Search for Energy-Aware NAS"],"prefix":"10.1109","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6745-9674","authenticated-orcid":false,"given":"Cairong","family":"Zhao","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology, Tongji University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3231-8817","authenticated-orcid":false,"given":"Shuguang","family":"Dou","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Tongji University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2805-8664","authenticated-orcid":false,"given":"Jiale","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Tongji University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4991-0596","authenticated-orcid":false,"given":"Xinyang","family":"Jiang","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia (Shanghai), Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-6527-8166","authenticated-orcid":false,"given":"Junyao","family":"Gao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Tongji University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-3095-3195","authenticated-orcid":false,"given":"Yuge","family":"Zhang","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia (Shanghai), Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8447-0928","authenticated-orcid":false,"given":"Bo","family":"Li","sequence":"additional","affiliation":[{"name":"S-Lab, Nanyang Technological University, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3103-8442","authenticated-orcid":false,"given":"Dongsheng","family":"Li","sequence":"additional","affiliation":[{"name":"Microsoft Research Asia (Shanghai), Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"doi-asserted-by":"publisher","key":"ref1","DOI":"10.1145\/3381831"},{"key":"ref2","article-title":"Full-cycle energy consumption benchmark for low-carbon computer vision","volume":"abs\/2108.13465","author":"Li","year":"2021","journal-title":"CoRR"},{"issue":"9","key":"ref3","first-page":"13693","article-title":"Energy and policy considerations for deep learning in NLP","volume-title":"Proc. AAAI Conf. Artif. Intell.","volume":"34","author":"Strubell","year":"2019"},{"doi-asserted-by":"publisher","key":"ref4","DOI":"10.1007\/978-3-030-05318-5_3"},{"doi-asserted-by":"publisher","key":"ref5","DOI":"10.1007\/978-3-030-05318-5_1"},{"year":"2009","author":"Krizhevsky","article-title":"Learning multiple layers of features from tiny images","key":"ref6"},{"doi-asserted-by":"publisher","key":"ref7","DOI":"10.1109\/CVPR.2019.00293"},{"doi-asserted-by":"publisher","key":"ref8","DOI":"10.1109\/CVPR.2019.01099"},{"doi-asserted-by":"publisher","key":"ref9","DOI":"10.1016\/j.knosys.2020.106622"},{"key":"ref10","article-title":"HW-NAS-bench: Hardware-aware neural architecture search benchmark","volume-title":"Proc. 9th Int. Conf. Learn. Representations","author":"Li","year":"2021"},{"key":"ref11","article-title":"Surrogate NAS benchmarks: Going beyond the limited search spaces of tabular NAS benchmarks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zela","year":"2022"},{"key":"ref12","first-page":"22534","article-title":"NAS-bench-x11 and the power of learning curves","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 34: Annu. Conf. Neural Inf. Process. Syst.","author":"Yan","year":"2021"},{"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."},{"year":"2023","article-title":"GPT-4 is OpenAI\u2019s most advanced system, producing safer and more useful responses","key":"ref14"},{"year":"2023","author":"Touvron","article-title":"Llama: Open and efficient foundation language models","key":"ref15"},{"doi-asserted-by":"publisher","key":"ref16","DOI":"10.18653\/v1\/2024.findings-acl.627"},{"doi-asserted-by":"publisher","key":"ref17","DOI":"10.18653\/v1\/2024.eacl-long.179"},{"key":"ref18","first-page":"4092","article-title":"Efficient neural architecture search via parameter sharing","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","volume":"80","author":"Pham","year":"2018"},{"key":"ref19","first-page":"5877","article-title":"The evolved transformer","volume-title":"Proc. 36th Int. Conf. Mach. Learn.","volume":"97","author":"So","year":"2019"},{"key":"ref20","first-page":"367","article-title":"Random search and reproducibility for neural architecture search","volume-title":"Proc. 35th Conf. Uncertainty Artif. Intell.","volume":"115","author":"Li","year":"2019"},{"key":"ref21","first-page":"7105","article-title":"Nas-bench-101: Towards reproducible neural architecture search","volume-title":"Proc. 36th Int. Conf. Mach. Learn.","volume":"97","author":"Ying","year":"2019"},{"key":"ref22","article-title":"NAS-bench-201: Extending the scope of reproducible neural architecture search","volume-title":"Proc. 8th Int. Conf. Learn. Representations","author":"Dong","year":"2020"},{"key":"ref23","article-title":"NAS-bench-suite: NAS evaluation is (Now) surprisingly easy","volume-title":"Proc. 10th Int. Conf. Learn. Representations","author":"Mehta","year":"2022"},{"year":"2020","author":"Dong","article-title":"Autohas: Efficient hyperparameter and architecture search","key":"ref24"},{"key":"ref25","article-title":"Tabular benchmarks for joint architecture and hyperparameter optimization","volume-title":"CoRR","volume":"abs\/1905.04970","author":"Klein","year":"2019"},{"key":"ref26","first-page":"1349","article-title":"NAS-HPO-bench-ii: A benchmark dataset on joint optimization of convolutional neural network architecture and training hyperparameters","volume-title":"Asian Conf. Mach. Learn.","volume":"157","author":"Hirose","year":"2021"},{"key":"ref27","first-page":"38788","article-title":"JAHS-bench-201: A foundation for research on joint architecture and hyperparameter search","volume-title":"Proc. 36th Conf. Neural Inf. Process. Syst. Datasets Benchmarks Track","author":"Bansal","year":"2022"},{"doi-asserted-by":"publisher","key":"ref28","DOI":"10.1109\/CVPR46437.2021.00938"},{"key":"ref29","first-page":"28904","article-title":"Stronger nas with weaker predictors","volume":"34","author":"Wu","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"doi-asserted-by":"publisher","key":"ref30","DOI":"10.1109\/TPAMI.2023.3293885"},{"key":"ref31","article-title":"Neural architecture search with reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zoph","year":"2017"},{"doi-asserted-by":"publisher","key":"ref32","DOI":"10.1609\/aaai.v33i01.33014780"},{"doi-asserted-by":"publisher","key":"ref33","DOI":"10.1145\/3579853"},{"doi-asserted-by":"publisher","key":"ref34","DOI":"10.1109\/MCSoC60832.2023.00034"},{"doi-asserted-by":"publisher","key":"ref35","DOI":"10.1109\/TPAMI.2021.3086914"},{"doi-asserted-by":"publisher","key":"ref36","DOI":"10.1109\/ISCAS51556.2021.9401187"},{"year":"2022","author":"Yu","article-title":"Resource-aware heterogeneous federated learning using neural architecture search","key":"ref37"},{"doi-asserted-by":"publisher","key":"ref38","DOI":"10.3390\/s21020444"},{"key":"ref39","first-page":"1","article-title":"Automl in the age of large language models: Current challenges, future opportunities and risks","author":"Tornede","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref40","article-title":"AutoML adoption in ML software","volume-title":"Proc. 8th ICML Workshop Automat. Mach. Learn.","author":"Blom","year":"2021"},{"doi-asserted-by":"publisher","key":"ref41","DOI":"10.1016\/j.mlwa.2020.100006"},{"doi-asserted-by":"publisher","key":"ref42","DOI":"10.1145\/3638529.3654017"},{"key":"ref43","first-page":"7787","article-title":"EvoPrompting: Language models for code-level neural architecture search","volume-title":"Proc. Adv. Neural Inf. Process. Systm","volume":"36","author":"Chen","year":"2023"},{"doi-asserted-by":"publisher","key":"ref44","DOI":"10.1002\/9781394217519.ch17"},{"key":"ref45","first-page":"28454","article-title":"How powerful are performance predictors in neural architecture search?","volume":"34","author":"White","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"year":"2023","author":"Yu","article-title":"GPT-NAS: Neural architecture search with the generative pre-trained model","key":"ref46"},{"doi-asserted-by":"publisher","key":"ref47","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref48","first-page":"185:1","article-title":"Hyperband: A novel bandit-based approach to hyperparameter optimization","volume":"18","author":"Li","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ref49","first-page":"3146","article-title":"LightGBM: A highly efficient gradient boosting decision tree","volume-title":"Adv. Neural Inf. Process. Syst. 30: Annu. Conf. Neural Inf. Process. Syst.","author":"Ke","year":"2017"},{"doi-asserted-by":"publisher","key":"ref50","DOI":"10.1109\/CVPR42600.2020.00983"},{"doi-asserted-by":"publisher","key":"ref51","DOI":"10.1609\/aaai.v29i1.9375"},{"issue":"7","key":"ref52","article-title":"Tiny imagenet visual recognition challenge","volume":"7","author":"Le","year":"2015","journal-title":"CS 231 N"},{"year":"2020","author":"Siems","article-title":"NAS-bench-301 and the case for surrogate benchmarks for neural architecture search","key":"ref53"},{"key":"ref54","article-title":"DARTS: Differentiable architecture search","volume-title":"Proc. 7th Int. Conf. Learn. Representations","author":"Liu","year":"2019"},{"key":"ref55","article-title":"NAS evaluation is frustratingly hard","volume-title":"Proc. 8th Int. Conf. Learn. Representations","author":"Yang","year":"2020"},{"key":"ref56","first-page":"3460","article-title":"Speeding up automatic hyperparameter optimization of deep neural networks by extrapolation of learning curves","volume-title":"Proc. 56th Int. Joint Conf. Artif. Intell.","author":"Domhan","year":"2015"},{"key":"ref57","first-page":"2546","article-title":"Algorithms for hyper-parameter optimization","volume-title":"Adv. Neural Inf. Process. Syst. 24: 25th Annu. Conf. Neural Inf. Process. Syst. 2011. Proc. A Meeting Held 12-14","author":"Bergstra","year":"2011"},{"doi-asserted-by":"publisher","key":"ref58","DOI":"10.1007\/978-3-030-58526-6_39"},{"key":"ref59","first-page":"28454","article-title":"How powerful are performance predictors in neural architecture search?","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 34: Annu. Conf. Neural Inf. Process. Syst.","author":"White","year":"2021"},{"doi-asserted-by":"publisher","key":"ref60","DOI":"10.1109\/ACCESS.2022.3169897"},{"key":"ref61","article-title":"Local search is state of the art for NAS benchmarks","volume":"abs\/2005.02960","author":"White","year":"2020","journal-title":"CoRR"},{"doi-asserted-by":"publisher","key":"ref62","DOI":"10.1609\/aaai.v35i12.17233"},{"key":"ref63","first-page":"1436","article-title":"BOHB: Robust and efficient hyperparameter optimization at scale","volume-title":"Proc. 35th Int. Conf. Mach. Learn., ICML, ser. Proc. Mach. Learn. Res.","volume":"80","author":"Falkner","year":"2018"},{"year":"2020","author":"Ruchte","article-title":"NASLib: A modular and flexible neural architecture search library","key":"ref64"},{"year":"2023","article-title":"GPT-3.5 turbo fine-tuning and api updates","key":"ref65"},{"year":"2023","article-title":"Gpt-4o contributions","key":"ref66"},{"year":"2024","author":"Hui","article-title":"Qwen2. 5-coder technical report","key":"ref67"},{"year":"2024","article-title":"Llama 3.2: Revolutionizing edge ai and vision with open, customizable models","key":"ref68"},{"key":"ref69","first-page":"24824","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref70","article-title":"Self-consistency improves chain of thought reasoning in language models","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Wang","year":"2023"},{"key":"ref71","first-page":"11809","article-title":"Tree of thoughts: Deliberate problem solving with large language models","volume":"36","author":"Yao","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11230086\/11122892.pdf?arnumber=11122892","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,6]],"date-time":"2025-11-06T18:53:24Z","timestamp":1762455204000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11122892\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12]]},"references-count":71,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3598206","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"type":"print","value":"0162-8828"},{"type":"electronic","value":"2160-9292"},{"type":"electronic","value":"1939-3539"}],"subject":[],"published":{"date-parts":[[2025,12]]}}}