publications
publications by categories in reversed chronological order. generated by jekyll-scholar.
2026
-
EVIDENT: Routing MLLM Adaptation through Entity-Grounded Visual Evidence for Cross-Domain Video Temporal GroundingarXiv preprint, 2026 -
Why Can’t I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action RecognitionIn ECCV, 2026 -
SlotVTG: Object-Centric Adapter for Generalizable Video Temporal GroundingIn GRAIL-V Workshop at CVPR, 2026
2024
-
DEVIAS: Learning Disentangled Video Representations of Action and Scene for Holistic Video UnderstandingIn ECCV (Oral, 2.3% acceptance rate), 2024