MungTso

LocVTP. [ECCV 22] LocVTP: Video-Text Pre-training for Temporal Localization

Continual-LLaVA.

DCNet. [ACM MM 22] Correspondence Matters for Video Referring Expression Comprehension

SpatialDreamer. SpatialDreamer: Incentivizing Spatial Reasoning via Active Mental Imagery

MILO. Seeing through Imagination: Learning Scene Geometry via Implicit Spatial World Modeling

Continual-RAG. Continual Retrieval-Augmented Generation for Large Vision-Language Models