datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
PittImageVideoAdsDataset
Dataset Card for PittImageVideoAdsDataset
Dataset Summary
PittImageVideoAdsDataset is the image and video advertisement dataset released with Automatic Understanding of Image and Video Advertisements. The paper reports 64,832 image advertisements and 3,477 YouTube advertisement videos, with human annotations for topics, sentiments, slogans, persuasive strategies, symbolic references, and action/reason Q/A. This Hugging Face version exposes the public annotation… See the full description on the dataset page: https://huggingface.co/datasets/creative-graphic-design/PittImageVideoAdsDataset.design-fto-bench
PatSnap Design FTO Bench
A Bench for evaluating design patent Freedom-To-Operate (FTO) retrieval systems on cross-modal image search. Each sample provides a query product image (or design patent figure) plus the ground truth set of target design patents that constitute infringement risk, as confirmed by patent invalidation proceedings.
🐙 GitHub mirror: This dataset is also published as part of the patsnap/patent-bench monorepo, where you can find the reference metric scripts… See the full description on the dataset page: https://huggingface.co/datasets/gigzjl/design-fto-bench.design-fto-bench
PatSnap Design FTO Bench
A Bench for evaluating design patent Freedom-To-Operate (FTO) retrieval systems on cross-modal image search. Each sample provides a query product image (or design patent figure) plus the ground truth set of target design patents that constitute infringement risk, as confirmed by patent invalidation proceedings.
🐙 GitHub mirror: This dataset is also published as part of the patsnap/patent-bench monorepo, where you can find the reference metric scripts… See the full description on the dataset page: https://huggingface.co/datasets/PatSnap/design-fto-bench.vibe-design-arena
Vibe Design Arena
The first public pairwise human preference dataset for real-world vibe-coded web applications.
60 apps from the Vibe Coding Showcase were compared pairwise by human annotators who judged which app has better visual design based on screenshots. Every possible pair was evaluated (C(60,2) = 1,770 comparisons), with 30 human votes per pair.
Dataset Summary
Stat
Value
Apps
60
Pairwise comparisons
1,770
Human votes per comparison
30
Total… See the full description on the dataset page: https://huggingface.co/datasets/datapointai/vibe-design-arena.