SHARP: Semantic Head-Aware Representation Pruning for Efficient MLLMs.

@inproceedings{ma-2026-sharp, author = {Haifeng Ma and Mingyue Guo and Linhui Xiao and Qingfang Zheng and Qingming Huang}, editor = {Stevan Rudinac and Zi Huang and Marcel Worring and Lucia Vadicamo and Xirong Li and Pascal Mettes and Lizi Liao}, title = {{SHARP: Semantic Head-Aware Representation Pruning for Efficient MLLMs}}, booktitle = {{Proceedings of the 2026 International Conference on Multimedia Retrieval, ICMR 2026, Amsterdam, The Netherlands, June 16-19, 2026}}, pages = {1701--1705}, publisher = {ACM}, year = {2026}, url = {https://doi.org/10.1145/3805622.3810733}, doi = {https://doi.org/10.1145/3805622.3810733} }