@inproceedings{mashvlmmitigatingactionscene, title = {MASH-VLM: Mitigating Action-Scene Hallucination in Video-LLMs through Disentangled Spatial-Temporal Representations}, author = {Kyungho Bae and Jinhyung Kim and Sihaeng Lee and Soonyoung Lee and GunHee Lee and Jinwoo Choi}, year = {2025}, booktitle = {CVPR 2025 1}, eprint = {2503.15871}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2503.15871v1}, }