VLM2Vec/DiDeMo

Clone from friedrichor/DiDeMo. About DiDeMo contains 10K long-form videos from Flickr. For each video, ~4 short sentences are annotated in temporal order. We follow the existing works to concatenate those short sentences and evaluate ‘paragraph-to-video’ retrieval on this benchmark. We adopt the official split: Train: 8,395 videos, 8,395 captions (concatenate from 33,005 short captions) Val: 1,065 videos, 1,065 captions (concatenate from 4,290 short captions) (We don't… See the full description on the dataset page: https://huggingface.co/datasets/VLM2Vec/DiDeMo.