Parameter-efficient modules

This paper introduces CREMA, a new and efficient modality-fusion framework designed to improve video reasoning. By leveraging existing pre-trained models, it incorporates multiple informative modalities from videos, such as optical…