TD-CoT: Bridging the Holistic-Atomic Gap for Training-Free Temporal Reversal Detection in Video LVLMs.

@inproceedings{huang-2026-tdcot, author = {Pengfei Huang and Xuezhen Hou}, editor = {Stevan Rudinac and Zi Huang and Marcel Worring and Lucia Vadicamo and Xirong Li and Pascal Mettes and Lizi Liao}, title = {{TD-CoT: Bridging the Holistic-Atomic Gap for Training-Free Temporal Reversal Detection in Video LVLMs}}, booktitle = {{Proceedings of the 2026 International Conference on Multimedia Retrieval, ICMR 2026, Amsterdam, The Netherlands, June 16-19, 2026}}, pages = {1542--1546}, publisher = {ACM}, year = {2026}, url = {https://doi.org/10.1145/3805622.3810743}, doi = {https://doi.org/10.1145/3805622.3810743} }