Dataset · Hugging Face

yaolily/GenS-Video-150K

Name: GenS-Video-150K
Creator: yaolily
License: apache-2.0
Keywords: task_categories:video-text-to-text, language:en, license:apache-2.0, size_categories:10K<n<100K, arxiv:2503.09146, region:us, frame-sampling, video-qa, video-llm, long-video-understanding, multimodal

🔗 Project Page · 📖 Paper · ⭐ GitHub · 📊 Dataset · 🤗 Checkpoints GenS-Video-150K Dataset To enable effective frame sampling, we introduce GenS-Video-150K, a large-scale synthetic dataset specifically designed for training frame sampling models. Annotated by GPT-4o, this dataset features: Dense coverage: Annotates ~20% of all frames with relevance scores. Fine-grained assessment: Assigns confidence scores (level 1 to 5) to relevant frames. Dataset Statistics… See the full description on the dataset page: https://huggingface.co/datasets/yaolily/GenS-Video-150K.

View on Hugging FaceReport an issue

Author / organization	yaolily
License (as declared)	apache-2.0
Downloads	212 (as reported by Hugging Face)
Likes	3
Created	13 March 2025
Last updated at source	4 September 2025

frame-samplingvideo-qavideo-llmlong-video-understandingmultimodal

Embed this entry

<iframe src="https://silytics.si/embed/item/5ad5a9ff-13eb-4f8a-8d6a-2b6006640be3" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 3 days

212 now (+11 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

Connected in the knowledge graph

yaolily/GenS-qwen2d5-vl-3bModel · used to train

Explore this entry in the graph

More from yaolily

yaolily/TimeChat-Online-139KDataset
yaolily/Timechat-OmniCaptioner-42KDataset
yaolily/OmniDCBenchDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.