LoRASculpt: Sculpting LoRA for Harmonizing General and Specialized Knowledge in Multimodal Large Language Models

@inproceedings{liang2025lorasculpt,
  title = {LoRASculpt: Sculpting LoRA for Harmonizing General and Specialized Knowledge in Multimodal Large Language Models},
  author = {Jian Liang and Wenke Huang and Guancheng Wan and Qu Yang and Mang Ye},
  year = {2025},
  booktitle = {Computer Vision and Pattern Recognition},
  eprint = {2503.16843},
  archivePrefix = {arXiv},
  doi = {10.1109/CVPR52734.2025.02437},
  url = {https://arxiv.org/abs/2503.16843},
}