Jul 2026
EVAS: Efficient Multimodal Temporal Forgery Localization via Audio-Visual Synergy and Steered Boundary Calibration
EVAS, an end-to-end multimodal framework for temporal forgery localization that adopts a decoupled training paradigm with auxiliary heads to disentangle representation learning from inference objectives, enhances model generalization and stability.
Shen Shen, Quan Zhang, Dan Jiang et al.
· arXiv.org · 0 citations