Co-occurring keywords · papers mentioning both
Papers
Augmenting Moment Retrieval: Zero-Dependency Two-Stage Learning
ICCV 2025
Text-guided 3D Human Generation from 2D Collections
EMNLP 2023
Exposing the Limits of Video-Text Models through Contrast Sets
NAACL 2022
Wolves at SemEval-2018 Task 10: Semantic Discrimination based on Knowledge and Association
SemEval 2018