CausalChapter: Improving Long-Video Chaptering with Interventional Dependency Modeling
CausalChapter is an intervention-inspired framework for long-video chaptering that estimates prediction-level influence through lightweight masking and removal interventions that improves boundary localization, chapter description quality, and cross-chapter coherence.
Xin-Ran Duan, Guo-Zhang Li, Yao-Yao Zhong et al.
· 0 citations