CommentsAccepted as a Spotlight at ICML 2026. The Version of Record will appear in Proceedings of Machine Learning Research (PMLR). 27 pages, 5 figures. Code: https://github.com/pixelhero98/LLapDiffusion
Leave it to the Specialist: Repair Sparse LLMs with Sparse Fine-Tuning via Sparsity Evolution
交给专家:通过稀疏性演化进行稀疏微调修复稀疏大语言模型
Qiao Xiao, Alan Ansell, Boqian Wu, Lu Yin, Mykola Pechenizkiy, Shiwei Liu, Decebal Constantin Mocanu
机构
*
Eindhoven University of Technology(埃因霍温理工大学)
;
University of Cambridge(剑桥大学)
;
University of Luxembourg(卢森堡大学)
;
University of Twente(埃因霍温理工大学)
;
University of Surrey(萨里大学)
;
Tübingen AI Center(图宾根人工智能中心)
;
Max Planck Institute for Intelligent Systems(智能系统马克斯·普朗克研究所)
;
ELLIS Institute Tübingen(图宾根ELLIS研究所)
Journal refProceedings of the Fifteenth Language Resources and Evaluation Conference (LREC 2026), pp. 4797-4818, European Language Resources Association (ELRA), Palma, Mallorca, Spain, May 2026