datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
JJokrobopoint-data
RoboPoint Dataset Card
Dataset details
This dataset contains 1432K image-QA instances used to fine-tune RoboPoint, a VLM for spatial affordance prediction. It consists of the following parts:
347K object reference instances from a synthetic data pipeline;
320K free space reference instances from a synthetic data pipeline;
100K object detection instaces from LVIS;
150K GPT-generated instruction-following instances from liuhaotian/LLaVA-Instruct-150K;
515K general-purpose… See the full description on the dataset page: https://huggingface.co/datasets/wentao-yuan/robopoint-data.fitvto-100k
FIT: A Large-Scale Dataset for Fit-Aware Virtual Try-On
The official preview dataset from the paper "FIT: A Large-Scale Dataset for Fit-Aware Virtual Try-On".
This dataset supports garment-centric virtual try-on and try-off research, containing 100,000 training and 5,000 evaluation triplets. Each sample pairs a person image with a layflat garment image and body/garment measurements.
Dataset Structure
Each split (train / eval) contains four aligned modalities — all… See the full description on the dataset page: https://huggingface.co/datasets/Yuanhao-Harry-Wang/fitvto-100k.ppv2_imageMip-NeRF360where2place
Where2Place Dataset Card
Dataset Details
This dataset contains 100 real-world images to evaluate free space reference using spatial relations. The images are collected from various cluttered environments. Each image is labeled with a sentence describing the desired some free space and a mask of the desired region.
Dataset Structure
images folder
Contains the raw images;
masks folder
Contains the corresponding binary masks for each image;… See the full description on the dataset page: https://huggingface.co/datasets/wentao-yuan/where2place.agent-reward-bench
AgentRewardBench
💾Code
📄Paper
🌐Website
🤗Dataset
💻Demo
🏆Leaderboard
AgentRewardBench: Evaluating Automatic Evaluations of Web Agent TrajectoriesXing Han Lù, Amirhossein Kazemnejad*, Nicholas Meade, Arkil Patel, Dongchan Shin, Alejandra Zambrano, Karolina Stańczak, Peter Shaw, Christopher J. Pal, Siva Reddy*Core Contributor
Loading dataset
You can use the huggingface_hub library to load the dataset. The dataset is available on Huggingface Hub at… See the full description on the dataset page: https://huggingface.co/datasets/yuanyyaa/agent-reward-bench.Subjects200K
Subjects200K is a large-scale dataset containing 200,000 paired images, introduced as part of the OminiControl project. Each image pair maintains subject consistency while presenting variations in scene context.
Quick Start
Load dataset
from datasets import load_dataset
# Load dataset
dataset = load_dataset('Yuanshi/Subjects200K')
Filter high-quality pairs from collection_2
def filter_func(item):
if item.get("collection") != "collection_2":
return False
if… See the full description on the dataset page: https://huggingface.co/datasets/Yuanshi/Subjects200K.ViViDoutput_claude-sonnet-4-6_0314yuag-numismatics
Yale University Art Gallery Numismatic Collection
This is a collection of over 53,000 coins held at the Yale University Art Gallery. The data were downloaded from Yale's Lux Collection Discovery. Lux let's users find and connect with the cultural heritage collections across Yale's museums, archives, and libraries in new ways and all in one place. The Numismatic Collection consist of over 70,000 objects. We filtered this dataset to only examples that had a single image. The dataset… See the full description on the dataset page: https://huggingface.co/datasets/yale-cultural-heritage/yuag-numismatics.qrcode_30kSubjects200K_collection3
Subjects200K is a large-scale dataset containing 200,000 paired images, introduced as part of the OminiControl project. Each image pair maintains subject consistency while presenting variations in scene context.
Quick Start
Load dataset
from datasets import load_dataset
# Load dataset
dataset = load_dataset('Yuanshi/Subjects200K')
Filter high-quality pairs from collection_2
def filter_func(item):
if item.get("collection") != "collection_2":
return False
if… See the full description on the dataset page: https://huggingface.co/datasets/Yuanshi/Subjects200K_collection3.Nurisk
Nurisk: VQA for Risk Assessment in Autonomous Driving
Nurisk is a visual question answering dataset focusing on risk assessment for autonomous driving. Each row contains:
image: a BEV image
question: a driving-related question
answer: the ground truth answer
Paper
NuRisk: A Visual Question Answering Dataset for Agent-Level Risk Assessment in Autonomous Driving — see the paper on arXiv:2509.25944 .
Framework
Dataset Structure… See the full description on the dataset page: https://huggingface.co/datasets/Yuan-avs/Nurisk.qrcode_image_100k_r16_textdrop0.5w3-imagesFIT-assetsbrainnetdreambench_plus
The image above shows the visualization of data distribution. (a) Images comparison between DreamBench and DreamBench++ using t-SNE. (b) Image and prompt distribution of DreamBench++.
DreamBench++ contains three categories: live subject (animals and humans), object, and style, with a total of 150 images. Among them, 120 images are photorealistic and 30 are non-photorealistic. Each image has 9 corresponding prompts, each with varying levels of difficulty, including 4 prompts for photorealistic… See the full description on the dataset page: https://huggingface.co/datasets/yuangpeng/dreambench_plus.IllumiCraft
IllumiCraft Dataset
This repository contains the dataset released with:
IllumiCraft: Unified Geometry and Illumination Diffusion for Controllable Video Generation
Yuanze Lin, Yi-Wen Chen, Yi-Hsuan Tsai, Ronald Clark, Ming-Hsuan Yang
🔗 Links
📄 Paper: https://arxiv.org/abs/2506.03150
🌐 Project Page: https://yuanze-lin.me/IllumiCraft_page/
💻 GitHub: https://github.com/yuanze-lin/IllumiCraft
🎥 YouTube: https://youtu.be/qAV58sADEzo
🤗 Checkpoints:… See the full description on the dataset page: https://huggingface.co/datasets/YuanzeLin/IllumiCraft.SynConFact
SynConFact
SynConFact is a synthetic counterfactual vision-language benchmark designed to evaluate how vision-language models respond to conflicts between visual information and prior knowledge in next-token prediction.
Dataset Description
SynConFact contains counterfactual image-text examples covering six categories:
Animal
Attribute
Historical
Material
Object
Scene
The images are synthetically generated using OpenAI image generation tools, while the textual… See the full description on the dataset page: https://huggingface.co/datasets/yuan1212/SynConFact.resembleOlympiadBench-OE-CoT-num10OpenECADv2-Datasets
OpenECADv2 Datasets
This repo releases data introducing in OpenECADv2 Core.
Dataset Name
Default View
Transparent View
Orthographic Views
100k_directout_1k
62500
25000
12500
150k_directout_2k
90000
37500
22500
200k_directout_3k
120000
50000
30000
These datasets are in LLaVA format.
Lora fine-tuning the OpenECADv2 0.89B model using OpenECADv2_100k_directout_1k only takes 2 hours on 2 x NVIDIA Geforce RTX 4090.
longvideogen_wavespeed_compact_3_trial3RGRecotoymicrowave_rgbdp_10fps_mu_15seprsi-synthetic-world-supervision-publicfonts_ds
Dataset Card for "fonts_ds"
More Information needed
fonts_en
Dataset Card for "fonts_en"
More Information needed