{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T06:27:54Z","timestamp":1774420074957,"version":"3.50.1"},"reference-count":53,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"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":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10889859","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T13:52:43Z","timestamp":1741787563000},"page":"1-5","source":"Crossref","is-referenced-by-count":0,"title":["GSMM: Efficient Global Sparsification for Resource-Conscious Multimodal Models"],"prefix":"10.1109","author":[{"given":"Wenlun","family":"Zhang","sequence":"first","affiliation":[{"name":"Keio University"}]},{"given":"Haoran","family":"Pang","sequence":"additional","affiliation":[{"name":"Keio University"}]},{"given":"Yucai","family":"Zhou","sequence":"additional","affiliation":[{"name":"Guangzhou Institute of Science and Technology"}]},{"given":"Shixiao","family":"Wang","sequence":"additional","affiliation":[{"name":"University of Birmingham"}]},{"given":"Luking","family":"Li","sequence":"additional","affiliation":[{"name":"Keio University"}]}],"member":"263","reference":[{"key":"ref1","article-title":"SliceGPT: Compress large language models by deleting rows and columns","volume-title":"The Twelfth International Conference on Learning Representations","author":"Ashkboos"},{"key":"ref2","article-title":"Gp-nas-ensemble: a model for the nas performance prediction","volume-title":"CVPRW","author":"Chen"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i15.33793"},{"key":"ref4","article-title":"Prunerzero: Evolving symbolic pruning metric from scratch for large language models","volume-title":"ICML","author":"Dong"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01145"},{"key":"ref6","article-title":"Stbllm: Breaking the 1-bit barrier with structured binary llms","author":"Dong","year":"2024"},{"key":"ref7","article-title":"Prior-guided one-shot neural architecture search","author":"Dong","year":"2022"},{"key":"ref8","article-title":"Sparsegpt: Massive language models can be accurately pruned in one-shot","author":"Frantar","year":"2023"},{"key":"ref9","article-title":"Sparsegpt: Massive language models can be accurately pruned in one-shot","author":"Frantar","year":"2023"},{"key":"ref10","article-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref12","article-title":"LoRA: Low-rank adaptation of large language models","volume-title":"International Conference on Learning Representations","author":"Hu"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2021.108025"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19809-0_20"},{"key":"ref15","article-title":"Detkds: Knowledge distillation search for object detectors","volume-title":"ICML","author":"Li"},{"key":"ref16","article-title":"Kd-zero: Evolving knowledge distiller for any teacher-student pairs","volume-title":"NeuIPS","author":"Li"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01597"},{"key":"ref18","article-title":"Shadow knowledge distillation: Bridging offline and online knowledge transfer","volume-title":"NeuIPS","author":"Li"},{"key":"ref19","article-title":"Discovering sparsity allocation for layer-wise pruning of large language models","volume-title":"NeuIPS","author":"Li"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN55064.2022.9891976"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/IJCNN55064.2022.9892575"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72652-1_3"},{"key":"ref23","article-title":"Explicit connection distillation","volume-title":"ICLR","author":"Li"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72652-1_2"},{"key":"ref25","article-title":"Als: Adaptive layer sparsity for large language models via activation correlation assessment","volume-title":"NeuIPS","author":"Li"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref27","article-title":"Not all patches are what you need: Expediting vision transformers via token reorganizations","author":"Liang","year":"2022"},{"key":"ref28","article-title":"Nora: Nested low-rank adaptation for efficient fine-tuning large models","author":"Lin","year":"2024"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02484"},{"key":"ref30","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref31","article-title":"Visual instruction tuning","author":"Liu","year":"2023"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i13.29327"},{"key":"ref34","article-title":"Learn to explain: Multimodal reasoning via thought chains for science question answering","author":"Lu","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref35","article-title":"Llm-pruner: On the structural pruning of large language models","author":"Ma","year":"2023"},{"key":"ref36","article-title":"Llm-pruner: On the structural pruning of large language models","author":"Ma","year":"2023"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-018-04316-3"},{"key":"ref38","volume-title":"OpenAI: Gpt-4 technical report","year":"2023"},{"key":"ref39","volume-title":"OpenAI: Gpt-4v(ision) system card","year":"2023"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i2.20108"},{"key":"ref41","article-title":"Catch-up distillation: You only need to train once for accelerating sampling","author":"Shao","year":"2023"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72652-1_4"},{"key":"ref44","article-title":"A simple and effective pruning approach for large language models","author":"Sun","year":"2023"},{"key":"ref45","article-title":"A simple and effective pruning approach for large language models","volume-title":"The Twelfth International Conference on Learning Representations","author":"Sun"},{"key":"ref46","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref47","author":"Xiaolong","year":"2022","journal-title":"Norm: Knowledge distillation via n-to-one representation matching"},{"key":"ref48","article-title":"BESA: Pruning large language models with blockwise parameter-efficient sparsity allocation","volume-title":"ICLR","author":"Xu"},{"key":"ref49","article-title":"Outlier weighed layerwise sparsity (owl): A missing secret sauce for pruning llms to high sparsity","author":"Yin","year":"2023"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i7.28606"},{"key":"ref51","article-title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023"},{"key":"ref52","article-title":"To prune, or not to prune: exploring the efficacy of pruning for model compression","author":"Zhu","year":"2017"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i14.29511"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Hyderabad, India","start":{"date-parts":[[2025,4,6]]},"end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10889859.pdf?arnumber=10889859","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T05:26:15Z","timestamp":1774416375000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10889859\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":53,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10889859","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}