The paper claims that a video-language model trained with dynamic temporal prompts and temporal contrastive learning beats four published models on three self-defined VidSitu temporal reasoning tasks.
Visual in-context le arning for large vision-language models,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Temporal Contrastive Learning for Video Temporal Reasoning in Large Vision-Language Models
The paper claims that a video-language model trained with dynamic temporal prompts and temporal contrastive learning beats four published models on three self-defined VidSitu temporal reasoning tasks.