Dataset · Hugging Face

FINEBENCH/FineBench

Name: FineBench
Creator: FINEBENCH
License: mit
Keywords: task_categories:visual-question-answering, task_categories:video-text-to-text, language:en, license:mit, size_categories:100K<n<1M, format:json, modality:image, modality:text, library:datasets, library:dask, library:mlcroissant, arxiv:2605.19846, region:us, CVPRW, VideoUnderstanding

Dataset Card for FineBench FineBench is a large-scale, multiple-choice Video Question Answering (VQA) dataset designed specifically to evaluate the fine-grained understanding of human actions in videos. It leverages the dense spatial (bounding boxes) and temporal (timestamps) annotations from the AVA v2.2 dataset, providing ~200k questions focused on nuanced person movements, interactions, and object manipulations within long video contexts. Dataset Details… See the full description on the dataset page: https://huggingface.co/datasets/FINEBENCH/FineBench.

View on Hugging FaceReport an issue

Author / organization	FINEBENCH
License (as declared)	mit
Downloads	2.4K (as reported by Hugging Face)
Likes	1
Created	7 April 2025
Last updated at source	23 May 2026

CVPRWVideoUnderstandingVideoQA

Embed this entry

<iframe src="https://silytics.si/embed/item/25153fec-16fc-4cfe-a882-6c73b6f488f2" 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

2,438 now (+9 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

Connected in the knowledge graph

FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity UnderstandingPaper · cites

Explore this entry in the graph

Related papers

FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity UnderstandingPaper

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 Information may be incomplete or out of date.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.