{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T18:38:17Z","timestamp":1792521497654,"version":"4.5.0"},"reference-count":44,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2027,1,1]],"date-time":"2027-01-01T00:00:00Z","timestamp":1798761600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62376228"],"award-info":[{"award-number":["62376228"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100019014","name":"Chengdu Science and Technology Program","doi-asserted-by":"publisher","award":["2025-YF12-00009-RC"],"award-info":[{"award-number":["2025-YF12-00009-RC"]}],"id":[{"id":"10.13039\/501100019014","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Neural Networks"],"published-print":{"date-parts":[[2027,1]]},"DOI":"10.1016\/j.neunet.2026.109558","type":"journal-article","created":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T14:03:54Z","timestamp":1787925834000},"page":"109558","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"PC","title":["Q-SA2: Enhanced low-bit quantization-aware fine-tuning via structure-aware adaptation"],"prefix":"10.1016","volume":"205","author":[{"given":"Yuning","family":"Yang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guowei","family":"Peng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3720-4379","authenticated-orcid":false,"given":"Xiurui","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4839-0234","authenticated-orcid":false,"given":"Dongyang","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guangchun","family":"Luo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guisong","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"3","key":"10.1016\/j.neunet.2026.109558_bib0001","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/1970392.1970395","article-title":"Robust principal component analysis?","volume":"58","author":"Cand\u00e8s","year":"2011","journal-title":"Journal of the ACM (JACM)"},{"issue":"240","key":"10.1016\/j.neunet.2026.109558_bib0002","first-page":"1","article-title":"PaLM: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"Journal of Machine Learning Research"},{"key":"10.1016\/j.neunet.2026.109558_bib0003","unstructured":"Cobbe, K., Kosaraju, V., Bavarian, M., Chen, M., Jun, H., Kaiser, L., Plappert, M., Tworek, J., Hilton, J., Nakano, R. et al. (2021), Training verifiers to solve math word problems, arXiv: 2110.14168."},{"key":"10.1016\/j.neunet.2026.109558_bib0004","doi-asserted-by":"crossref","first-page":"30318","DOI":"10.52202\/068431-2198","article-title":"LLM.int8: 8-bit matrix multiplication for transformers at scale","volume":"35","author":"Dettmers","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0005","doi-asserted-by":"crossref","first-page":"10088","DOI":"10.52202\/075280-0441","article-title":"QLoRA: Efficient finetuning of quantized llms","volume":"36","author":"Dettmers","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0006","doi-asserted-by":"crossref","unstructured":"Ding, S., Meadowlark, P., He, Y., Lew, L., Agrawal, S., & Rybakov, O. (2022), 4-bit conformer with native quantization aware training for speech recognition, arXiv: 2203.15952.","DOI":"10.21437\/Interspeech.2022-10809"},{"key":"10.1016\/j.neunet.2026.109558_bib0007","series-title":"Proceedings of the IEEE\/CVF international conference on computer vision","first-page":"293","article-title":"HAWQ: Hessian aware quantization of neural networks with mixed-precision","author":"Dong","year":"2019"},{"key":"10.1016\/j.neunet.2026.109558_bib0008","series-title":"The eleventh international conference on learning representations","article-title":"OPTQ: Accurate quantization for generative pre-trained transformers","author":"Frantar","year":"2023"},{"key":"10.1016\/j.neunet.2026.109558_bib0009","series-title":"Low-power computer vision","first-page":"291","article-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2022"},{"key":"10.1016\/j.neunet.2026.109558_bib0010","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2025.107856","article-title":"A survey of low-bit large language models: Basics, systems, and algorithms","volume":"192","author":"Gong","year":"2025","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109558_bib0011","series-title":"Findings of the association for computational linguistics: ACL 2023","first-page":"8065","article-title":"PreQuant: A task-agnostic quantization approach for pre-trained language models","author":"Gong","year":"2023"},{"key":"10.1016\/j.neunet.2026.109558_bib0012","unstructured":"Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Vaughan, A. et al. (2024), The llama 3 herd of models, arXiv: 2407.21783."},{"key":"10.1016\/j.neunet.2026.109558_bib0013","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2025.108095","article-title":"La-LoRA: Parameter-efficient fine-tuning with layer-wise adaptive low-rank adaptation","volume":"194","author":"Gu","year":"2026","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109558_bib0014","series-title":"The twelfth international conference on learning representations","article-title":"LQ-LoRA: Low-rank plus quantized matrix decomposition for efficient language model finetuning","author":"Guo","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0015","series-title":"The eleventh international conference on learning representations","article-title":"DeBERTaV3: Improving DeBERTa using ELECTRA-style pre-training with gradient-disentangled embedding sharing","author":"He","year":"2023"},{"key":"10.1016\/j.neunet.2026.109558_bib0016","series-title":"The tenth international conference on learning representations","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2022"},{"key":"10.1016\/j.neunet.2026.109558_bib0017","unstructured":"Hui, B., Yang, J., Cui, Z., Yang, J., Liu, D., Zhang, L., Liu, T., Zhang, J., Yu, B., Lu, K. et al. (2024), Qwen2.5-coder technical report, arXiv: 2409.12186."},{"key":"10.1016\/j.neunet.2026.109558_bib0018","first-page":"36187","article-title":"Memory-efficient fine-tuning of compressed large language models via sub-4-bit integer quantization","volume":"36","author":"Kim","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0019","series-title":"Proceedings of the 2022 conference on empirical methods in natural language processing","first-page":"6713","article-title":"Understanding and improving knowledge distillation for quantization aware training of large transformer encoders","author":"Kim","year":"2022"},{"key":"10.1016\/j.neunet.2026.109558_bib0020","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106414","article-title":"Hydra: Multi-head low-rank adaptation for parameter efficient fine-tuning","volume":"178","author":"Kim","year":"2024","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109558_bib0021","series-title":"The twelfth international conference on learning representations","article-title":"NOLA: Compressing LoRA using linear combination of random basis","author":"Koohpayegani","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0022","series-title":"Proceedings of the AAAI conference on artificial intelligence","first-page":"13355","article-title":"OWQ: Outlier-aware weight quantization for efficient fine-tuning and inference of large language models","author":"Lee","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0023","series-title":"The twelfth international conference on learning representations","article-title":"LoftQ: LoRA-fine-tuning-aware quantization for large language models","author":"Li","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0024","series-title":"Proceedings of the 2024 conference on empirical methods in natural language processing","first-page":"20996","article-title":"ApiQ: Finetuning of 2-bit quantized large language model","author":"Liao","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0025","first-page":"15186","article-title":"Make pre-trained model reversible: From parameter to memory efficient fine-tuning","volume":"36","author":"Liao","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0026","first-page":"87","article-title":"AWQ: Activation-aware weight quantization for on-device llm compression and acceleration","volume":"6","author":"Lin","year":"2024","journal-title":"Proceedings of Machine Learning and Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0027","series-title":"2nd workshop on advancing neural network training: Computational efficiency, scalability, and resource optimizatio","article-title":"SVFT: Parameter-efficient fine-tuning with singular vectors","author":"Lingam","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0028","doi-asserted-by":"crossref","first-page":"13579","DOI":"10.52202\/079017-0434","article-title":"BitDelta: Your fine-tune may only be worth one bit","volume":"37","author":"Liu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0029","series-title":"Findings of the association for computational linguistics: ACL 2024","first-page":"467","article-title":"LLM-QAT: Data-free quantization aware training for large language models","author":"Liu","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0030","doi-asserted-by":"crossref","first-page":"121038","DOI":"10.52202\/079017-3846","article-title":"PiSSA: Principal singular values and singular vectors adaptation of large language models","volume":"37","author":"Meng","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0031","unstructured":"Merity, S., Xiong, C., Bradbury, J., & Socher, R. (2016), Pointer sentinel mixture models, arXiv: 1609.07843."},{"key":"10.1016\/j.neunet.2026.109558_bib0032","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109558_bib0033","doi-asserted-by":"crossref","first-page":"31056","DOI":"10.52202\/079017-0978","article-title":"Delta-CoMe: Training-free delta-compression with mixed-precision for large language models","volume":"37","author":"Ping","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"140","key":"10.1016\/j.neunet.2026.109558_bib0034","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":"10.1016\/j.neunet.2026.109558_bib0035","series-title":"The Twelfth international conference on learning representations","article-title":"OmniQuant: Omnidirectionally calibrated quantization for large language models","author":"Shao","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0036","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2025.107909","article-title":"LS-PRISM: A layer-selective pruning method via low-rank approximation and sparsification for efficient large language model compression","volume":"192","author":"Tao","year":"2025","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109558_bib0037","unstructured":"G. Team, Mesnard, T., Hardin, C., Dadashi, R., Bhupatiraju, S., Pathak, S., Sifre, L., Rivi\u00e8re, M., Kale, M. S., Love, J. et al. (2024), Gemma: Open models based on gemini research and technology, arXiv: 2403.08295."},{"key":"10.1016\/j.neunet.2026.109558_bib0038","unstructured":"Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S. et al. (2023), Llama 2: Open foundation and fine-tuned chat models, arXiv: 2307.09288."},{"key":"10.1016\/j.neunet.2026.109558_bib0039","doi-asserted-by":"crossref","unstructured":"Wang, A. (2018), GLUE: A multi-task benchmark and analysis platform for natural language understanding, arXiv: 1804.07461.","DOI":"10.18653\/v1\/W18-5446"},{"key":"10.1016\/j.neunet.2026.109558_bib0040","unstructured":"Workshop, B., Scao, T. L., Fan, A., Akiki, C., Pavlick, E., Ili\u0107, S., Hesslow, D., Castagn\u00e9, R., Luccioni, A. S., Yvon, F. et al. (2022), BLOOM: A 176b-parameter open-access multilingual language model, arXiv: 2211.05100."},{"key":"10.1016\/j.neunet.2026.109558_bib0041","series-title":"International conference on machine learning","first-page":"38087","article-title":"SmoothQuant: Accurate and efficient post-training quantization for large language models","author":"Xiao","year":"2023"},{"key":"10.1016\/j.neunet.2026.109558_bib0042","series-title":"Forty-first international conference on machine learning","article-title":"LQER: Low-rank quantization error reconstruction for LLMs","author":"Zhang","year":"2024"},{"key":"10.1016\/j.neunet.2026.109558_bib0043","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Zhang, S., Cao, S., Du, D., Wei, J., Cao, T., & Xu, N. (2023), AFPQ: Asymmetric floating point quantization for llms, arXiv: 2311.01792.","DOI":"10.18653\/v1\/2024.findings-acl.3"},{"key":"10.1016\/j.neunet.2026.109558_bib0044","doi-asserted-by":"crossref","first-page":"1556","DOI":"10.1162\/tacl_a_00704","article-title":"A survey on model compression for large language models","volume":"12","author":"Zhu","year":"2024","journal-title":"Transactions of the Association for Computational Linguistics"}],"container-title":["Neural Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026010166?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026010166?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T17:38:14Z","timestamp":1792517894000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0893608026010166"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2027,1]]},"references-count":44,"alternative-id":["S0893608026010166"],"URL":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109558","relation":{},"ISSN":["0893-6080"],"issn-type":[{"value":"0893-6080","type":"print"}],"subject":[],"published":{"date-parts":[[2027,1]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Q-SA2: Enhanced low-bit quantization-aware fine-tuning via structure-aware adaptation","name":"articletitle","label":"Article Title"},{"value":"Neural Networks","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109558","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"109558"}}