VSTAR: A Video-grounded Dialogue Dataset for Situated Semantic Understanding with Scene and Topic Transitions
38578Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
VSTAR
Material |
isBasedOn |
Генерация диалога
Task |
|
VSTAR
Material |
isBasedOn |
Сегментация сцены
Task |
|
VSTAR
Material |
isBasedOn |
Сегментация тем
Method |