{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T21:28:25Z","timestamp":1786483705379,"version":"3.56.0"},"reference-count":51,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"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":["62536004"],"award-info":[{"award-number":["62536004"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Neurocomputing"],"published-print":{"date-parts":[[2026,10]]},"DOI":"10.1016\/j.neucom.2026.134299","type":"journal-article","created":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T16:24:33Z","timestamp":1781627073000},"page":"134299","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["A coarse-to-fine dynamic layer pruning framework for parameter-efficient fine-tuning"],"prefix":"10.1016","volume":"698","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-0806-2908","authenticated-orcid":false,"given":"Xin","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6847-6740","authenticated-orcid":false,"given":"Shuzhen","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4145-823X","authenticated-orcid":false,"given":"Zhulin","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C.L. Philip","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.neucom.2026.134299_bib0005","series-title":"ACL","first-page":"7319","article-title":"Intrinsic dimensionality explains the effectiveness of language model fine-tuning","author":"Aghajanyan","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0010","series-title":"ICLR","article-title":"DeBERTaV3: improving DeBERTa using ELECTRA-style pre-training with gradient-disentangled embedding sharing","author":"He","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0015","first-page":":140:1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.neucom.2026.134299_bib0020","series-title":"LLaMA: open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0030","series-title":"Proceedings of the 36th International Conference on Machine Learning, 97","first-page":"2790","article-title":"Parameter-efficient transfer learning for NLP","author":"Houlsby","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0035","series-title":"ICLR","article-title":"LoRA: low-rank adaptation of large language models","author":"Hu","year":"2022"},{"key":"10.1016\/j.neucom.2026.134299_bib0040","doi-asserted-by":"crossref","first-page":"4253","DOI":"10.1109\/TASLP.2024.3463395","article-title":"Bayesian parameter-efficient fine-tuning for overcoming catastrophic forgetting","volume":"32","author":"Chen","year":"2024","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"10.1016\/j.neucom.2026.134299_bib0045","doi-asserted-by":"crossref","first-page":"3002","DOI":"10.1109\/TASLP.2024.3407519","article-title":"Black-box prompt tuning with subspace learning","volume":"32","author":"Zheng","year":"2024","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"10.1016\/j.neucom.2026.134299_bib0050","doi-asserted-by":"crossref","first-page":"1061","DOI":"10.1109\/TASLP.2023.3267618","article-title":"Can pretrained english language models benefit non-english NLP systems in low-resource scenarios?","volume":"32","author":"Chi","year":"2024","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"10.1016\/j.neucom.2026.134299_bib0055","doi-asserted-by":"crossref","first-page":"4607","DOI":"10.1109\/TASLP.2024.3477330","article-title":"Derivative-free optimization for low-rank adaptation in large language models","volume":"32","author":"Jin","year":"2024","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"10.1016\/j.neucom.2026.134299_bib0060","doi-asserted-by":"crossref","first-page":"3867","DOI":"10.1109\/TASLP.2024.3434445","article-title":"ELP-adapters: parameter efficient adapter tuning for various speech processing tasks","volume":"32","author":"Inoue","year":"2024","journal-title":"IEEE\/ACM Transactions on Audio, Speech, and Language Processing"},{"key":"10.1016\/j.neucom.2026.134299_bib0065","series-title":"EMNLP","first-page":"7930","article-title":"AdapterDrop: on the efficiency of adapters in transformers","author":"R\u00fcckl\u00e9","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0070","series-title":"Findings of the Association for Computational Linguistics: EMNLP","first-page":"3105","article-title":"On surgical fine-tuning for language encoders","author":"Lodha","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0075","series-title":"NeurIPS","article-title":"LISA: layerwise importance sampling for memory-efficient large language model fine-tuning","author":"Pan","year":"2024"},{"key":"10.1016\/j.neucom.2026.134299_bib0080","series-title":"ICLR","article-title":"Intermediate layer classifiers for OOD generalization","author":"Uselis","year":"2025"},{"key":"10.1016\/j.neucom.2026.134299_bib0085","series-title":"Advances in Neural Information Processing Systems","first-page":"24193","article-title":"Training neural networks with fixed sparse masks","author":"Sung","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0090","series-title":"ICLR","article-title":"SmartFRZ: an efficient training framework using attention-based layer freezing","author":"Li","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0095","series-title":"ICLR","article-title":"Adaptive budget allocation for parameter-efficient fine-tuning","author":"Zhang","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0100","series-title":"ICLR","article-title":"Budgeted online continual learning by adaptive layer freezing and frequency-based sampling","author":"Seo","year":"2025"},{"key":"10.1016\/j.neucom.2026.134299_bib0105","series-title":"The Eleventh International Conference on Learning Representations","article-title":"Surgical fine-tuning improves adaptation to distribution shifts","author":"Lee","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0110","series-title":"Advances in Neural Information Processing Systems, 35","first-page":"19061","article-title":"Layer freezing & data sieving: missing pieces of a generic framework for sparse training","author":"Yuan","year":"2022"},{"key":"10.1016\/j.neucom.2026.134299_bib0115","series-title":"Proceedings of the 31st International Conference on Computational Linguistics","first-page":"5530","article-title":"LoRA-drop: efficient LoRA parameter pruning based on output evaluation","author":"Zhou","year":"2025"},{"key":"10.1016\/j.neucom.2026.134299_bib0120","series-title":"NeurIPS, 35","first-page":"26462","article-title":"ST-adapter: parameter-efficient image-to-video transfer learning","author":"Pan","year":"2022"},{"issue":"17","key":"10.1016\/j.neucom.2026.134299_bib0125","doi-asserted-by":"crossref","first-page":"18824","DOI":"10.1609\/aaai.v38i17.29847","article-title":"READ-PVLA: recurrent adapter with partial video-language alignment for parameter-efficient transfer learning in low-resource video-language modeling","volume":"38","author":"Nguyen","year":"2024","journal-title":"AAAI"},{"key":"10.1016\/j.neucom.2026.134299_bib0130","series-title":"EMNLP","first-page":"3045","article-title":"The power of scale for parameter-efficient prompt tuning","author":"Lester","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0135","series-title":"ACL","first-page":"4582","article-title":"Prefix-tuning: optimizing continuous prompts for generation","author":"Li","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0140","series-title":"EMNLP","first-page":"14306","article-title":"SMoP: towards efficient and effective prompt tuning with sparse mixture-of-prompts","author":"Choi","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0145","series-title":"NeurIPS, 36","first-page":"10088","article-title":"QLoRA: efficient finetuning of quantized LLMs","author":"Dettmers","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0150","series-title":"ICML","article-title":"DoRA: weight-decomposed low-rank adaptation","author":"yang Liu","year":"2024"},{"key":"10.1016\/j.neucom.2026.134299_bib0155","series-title":"The Thirteenth International Conference on Learning Representations","article-title":"RandLoRA: full rank parameter-efficient fine-tuning of large models","author":"Albert","year":"2025"},{"key":"10.1016\/j.neucom.2026.134299_bib0160","series-title":"The Thirteenth International Conference on Learning Representations","article-title":"HiRA: parameter-efficient hadamard high-rank adaptation for large language models","author":"Huang","year":"2025"},{"issue":"1","key":"10.1016\/j.neucom.2026.134299_bib0165","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1109\/TNNLS.2017.2716952","article-title":"Broad learning system: an effective and efficient incremental learning system without the need for deep architecture","volume":"29","author":"Chen","year":"2018","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"issue":"9","key":"10.1016\/j.neucom.2026.134299_bib0170","doi-asserted-by":"crossref","first-page":"12731","DOI":"10.1109\/TNNLS.2023.3264617","article-title":"A broad generative network for two-stage image outpainting","volume":"35","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"issue":"7","key":"10.1016\/j.neucom.2026.134299_bib0175","doi-asserted-by":"crossref","first-page":"6232","DOI":"10.1109\/TCYB.2021.3050508","article-title":"BMT-net: broad multitask transformer network for sentiment analysis","volume":"52","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Cybern."},{"key":"10.1016\/j.neucom.2026.134299_bib0180","series-title":"Naacl-Hlt","first-page":"4171","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0185","first-page":"62","article-title":"Greedy algorithm","volume":"2","author":"Black","year":"2005","journal-title":"Dictionary of Algorithms and Data Structures"},{"key":"10.1016\/j.neucom.2026.134299_bib0190","series-title":"NeurIPS, 34","first-page":"24193","article-title":"Training neural networks with fixed sparse masks","author":"Sung","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0195","series-title":"CVPR","first-page":"11256","article-title":"Importance estimation for neural network pruning","author":"Molchanov","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0200","series-title":"Proceedings of the 39th International Conference on Machine Learning, 162","first-page":"26809","article-title":"PLATON: pruning large transformer models with upper confidence bound of weight importance","author":"Zhang","year":"2022"},{"key":"10.1016\/j.neucom.2026.134299_bib0210","series-title":"ICLR","article-title":"GLUE: a multi-task benchmark and analysis platform for natural language understanding","author":"Wang","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0215","series-title":"NAACL","first-page":"2924","article-title":"BoolQ: exploring the surprising difficulty of natural Yes\/No questions","author":"Clark","year":"2019"},{"issue":"5","key":"10.1016\/j.neucom.2026.134299_bib0220","doi-asserted-by":"crossref","first-page":"7432","DOI":"10.1609\/aaai.v34i05.6239","article-title":"PIQA: reasoning about physical commonsense in natural language","volume":"34","author":"Bisk","year":"2020","journal-title":"AAAI"},{"key":"10.1016\/j.neucom.2026.134299_bib0225","series-title":"EMNLP-IJCNLP","first-page":"4463","article-title":"Social IQa: commonsense reasoning about social interactions","author":"Sap","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0230","series-title":"ACL","first-page":"4791","article-title":"HellaSwag: can a machine really finish your sentence?","author":"Zellers","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0235","series-title":"ICLR","article-title":"Measuring massive multitask language understanding","author":"Hendrycks","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0240","series-title":"Advances in Neural Information Processing Systems","first-page":"8024","article-title":"PyTorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"key":"10.1016\/j.neucom.2026.134299_bib0245","series-title":"Advances in Neural Information Processing Systems","first-page":"1022","article-title":"Compacter: efficient low-rank hypercomplex adapter layers","author":"Mahabadi","year":"2021"},{"key":"10.1016\/j.neucom.2026.134299_bib0250","series-title":"EMNLP","first-page":"5254","article-title":"LLM-adapters: an adapter family for parameter-efficient fine-tuning of large language models","author":"Hu","year":"2023"},{"key":"10.1016\/j.neucom.2026.134299_bib0255","series-title":"EMNLP","first-page":"38","article-title":"Transformers: state-of-the-art natural language processing","author":"Wolf","year":"2020"},{"key":"10.1016\/j.neucom.2026.134299_bib0260","series-title":"ACL","first-page":"1","article-title":"BitFit: simple parameter-efficient fine-tuning for transformer-based masked language-models","author":"Zaken","year":"2022"},{"key":"10.1016\/j.neucom.2026.134299_bib0265","doi-asserted-by":"crossref","DOI":"10.1016\/j.neucom.2025.131913","article-title":"When less is more: sample-aware pruning of uninformative features improves neural-network regression","volume":"660","author":"Christakis","year":"2026","journal-title":"Neurocomputing"}],"container-title":["Neurocomputing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0925231226016978?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0925231226016978?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T20:37:02Z","timestamp":1786480622000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0925231226016978"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":51,"alternative-id":["S0925231226016978"],"URL":"https:\/\/doi.org\/10.1016\/j.neucom.2026.134299","relation":{},"ISSN":["0925-2312"],"issn-type":[{"value":"0925-2312","type":"print"}],"subject":[],"published":{"date-parts":[[2026,10]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"A coarse-to-fine dynamic layer pruning framework for parameter-efficient fine-tuning","name":"articletitle","label":"Article Title"},{"value":"Neurocomputing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neucom.2026.134299","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Published by Elsevier B.V.","name":"copyright","label":"Copyright"}],"article-number":"134299"}}