@inproceedings{boltboostlargevisionlanguagemodel, title = {BOLT: Boost Large Vision-Language Model Without Training for Long-form Video Understanding}, author = {Shuming Liu and Chen Zhao and Tianqi Xu and Bernard Ghanem}, year = {2025}, booktitle = {CVPR 2025 1}, eprint = {2503.21483}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2503.21483v1}, }