Song-Hua Liu

This paper proposes SVEET, a framework that requires merely training on a pretrained bidirectional video diffusion model but supports high-quality streaming video editing in an auto-regressive fashion and proposes a decoupled training scheme that explicitly enforces the orthogonality between the optimization directions of video controllability and model causality.

Action-class miscalibration is a hidden failure mode that can significantly skew performance evaluations of LLMs in multi-turn tool-calling tasks.