Explainable Forensics of Manipulated Segments in Untrimmed Long Videos

Yue Feng, Jingjing Li, Qijia Lu, Wei Ji, Jingrou Zhang, Fei Shen, Xiao Li, Yizhen Jia, Qiang Chen, Limin Wang, Wentong Li, Jie Qin
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:30115-30131, 2026.

Abstract

The rapid advancement of AI-driven video generation has transformed content creation, while simultaneously increasing the risk of misinformation through localized manipulations in long-form videos. Existing video forensic methods predominantly operate on short, independent clips, and thus fail to capture realistic scenarios where AI-generated content is sparsely embedded within otherwise authentic footage. To bridge this gap, we formulate the task of Temporal AI-Generated Segment Localization and Explanation, which targets authenticity detection, temporal localization, and interpretable analysis of manipulated segments in untrimmed long videos. We further introduce TASLE, a large-scale benchmark comprising 12,472 untrimmed videos with diverse manipulation patterns and rich annotation signals, including temporal boundaries, authenticity labels, and segment-level rationales. In addition, we propose MSLoc, a coarse-to-fine forensic baseline that combines a boundary-sensitive proposal generation module for efficient long-video scanning with an MLLM-based refinement module for precise boundary localization and interpretable reasoning. Experiments validate the effectiveness of the proposed baseline, highlighting the importance of segment-level explainable forensics for long-form AI-generated video analysis. Dataset and code will be made publicly available.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-feng26g, title = {Explainable Forensics of Manipulated Segments in Untrimmed Long Videos}, author = {Feng, Yue and Li, Jingjing and Lu, Qijia and Ji, Wei and Zhang, Jingrou and Shen, Fei and Li, Xiao and Jia, Yizhen and Chen, Qiang and Wang, Limin and Li, Wentong and Qin, Jie}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {30115--30131}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/feng26g/feng26g.pdf}, url = {https://proceedings.mlr.press/v306/feng26g.html}, abstract = {The rapid advancement of AI-driven video generation has transformed content creation, while simultaneously increasing the risk of misinformation through localized manipulations in long-form videos. Existing video forensic methods predominantly operate on short, independent clips, and thus fail to capture realistic scenarios where AI-generated content is sparsely embedded within otherwise authentic footage. To bridge this gap, we formulate the task of Temporal AI-Generated Segment Localization and Explanation, which targets authenticity detection, temporal localization, and interpretable analysis of manipulated segments in untrimmed long videos. We further introduce TASLE, a large-scale benchmark comprising 12,472 untrimmed videos with diverse manipulation patterns and rich annotation signals, including temporal boundaries, authenticity labels, and segment-level rationales. In addition, we propose MSLoc, a coarse-to-fine forensic baseline that combines a boundary-sensitive proposal generation module for efficient long-video scanning with an MLLM-based refinement module for precise boundary localization and interpretable reasoning. Experiments validate the effectiveness of the proposed baseline, highlighting the importance of segment-level explainable forensics for long-form AI-generated video analysis. Dataset and code will be made publicly available.} }
Endnote
%0 Conference Paper %T Explainable Forensics of Manipulated Segments in Untrimmed Long Videos %A Yue Feng %A Jingjing Li %A Qijia Lu %A Wei Ji %A Jingrou Zhang %A Fei Shen %A Xiao Li %A Yizhen Jia %A Qiang Chen %A Limin Wang %A Wentong Li %A Jie Qin %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-feng26g %I PMLR %P 30115--30131 %U https://proceedings.mlr.press/v306/feng26g.html %V 306 %X The rapid advancement of AI-driven video generation has transformed content creation, while simultaneously increasing the risk of misinformation through localized manipulations in long-form videos. Existing video forensic methods predominantly operate on short, independent clips, and thus fail to capture realistic scenarios where AI-generated content is sparsely embedded within otherwise authentic footage. To bridge this gap, we formulate the task of Temporal AI-Generated Segment Localization and Explanation, which targets authenticity detection, temporal localization, and interpretable analysis of manipulated segments in untrimmed long videos. We further introduce TASLE, a large-scale benchmark comprising 12,472 untrimmed videos with diverse manipulation patterns and rich annotation signals, including temporal boundaries, authenticity labels, and segment-level rationales. In addition, we propose MSLoc, a coarse-to-fine forensic baseline that combines a boundary-sensitive proposal generation module for efficient long-video scanning with an MLLM-based refinement module for precise boundary localization and interpretable reasoning. Experiments validate the effectiveness of the proposed baseline, highlighting the importance of segment-level explainable forensics for long-form AI-generated video analysis. Dataset and code will be made publicly available.
APA
Feng, Y., Li, J., Lu, Q., Ji, W., Zhang, J., Shen, F., Li, X., Jia, Y., Chen, Q., Wang, L., Li, W. & Qin, J.. (2026). Explainable Forensics of Manipulated Segments in Untrimmed Long Videos. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:30115-30131 Available from https://proceedings.mlr.press/v306/feng26g.html.

Related Material