TSAD: Trace-Semantic Adaptive Disentanglement for Detecting and Grounding Multi-Modal Media Manipulation.

@inproceedings{liu-2026-tsad, author = {Wenzheng Liu and Cheng Fu and Hujin Peng and Junlong Wu and Xianhong Chen and Lan Huang and Jing Xu and Yixi Tian and Menghan Liang and Xiaofeng Wang and Tan Deng and Wanwei Jiang and Xiang Li}, editor = {Stevan Rudinac and Zi Huang and Marcel Worring and Lucia Vadicamo and Xirong Li and Pascal Mettes and Lizi Liao}, title = {{TSAD: Trace-Semantic Adaptive Disentanglement for Detecting and Grounding Multi-Modal Media Manipulation}}, booktitle = {{Proceedings of the 2026 International Conference on Multimedia Retrieval, ICMR 2026, Amsterdam, The Netherlands, June 16-19, 2026}}, pages = {2077--2086}, publisher = {ACM}, year = {2026}, url = {https://doi.org/10.1145/3805622.3810582}, doi = {https://doi.org/10.1145/3805622.3810582} }