datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
imagenet_hard_review_data_r2imagenet_hard_review_dataimagenet_sketch_resizedimagenet-metric-refsimagenet-metric-refsCSFM-ImageNet1K-Caption
CSFM-ImageNet1K-Caption Dataset
Project Page | Paper | Code
This repository contains dataset associated with the paper "Better Source Better Flow: Learning Condition-Dependent Source Distribution for Flow Matching".
This dataset is used for training and evaluating Condition-dependent Source Flow Matching (CSFM), a framework that learns condition-dependent source distributions for flow matching. We recaptioned the ImageNet-1K dataset using Qwen3-VL-8B Instruct, resulting in detailed… See the full description on the dataset page: https://huggingface.co/datasets/junwann/CSFM-ImageNet1K-Caption.hello-ot-imagenet-pca
HELLO representative ImageNet VA-VAE PCA features
This dataset contains four independently generated float32 NumPy matrices,
with shapes (262144, d) for d = 4, 32, 256, 2048, totaling about 2.29 GiB.
All use seed=42 and serve the public main-scaling and parameter-sensitivity
experiments. Larger sample counts are outside the published data scope.
The artifact contains numeric PCA-projected features only. It contains no
images, labels, captions, filenames, or ImageNet identifiers.… See the full description on the dataset page: https://huggingface.co/datasets/WenzhouXia/hello-ot-imagenet-pca.imagenet-variations-synth-pilot-v3
ImageNet Variations Synth Pilot v3.1 (diverse, 1K)
Huu + Claude multimodal instruction pipeline — quality-fixed re-run.
Pipeline
Flux.1-schnell (Huu style phrasings + visual style boosts + aspect ratios)
→ Seed2 → Florence-2 grounding/caption → tracks A–F → USER / ASSISTANT.
Tracks
A programmatic QA (counts/style/spatial/absence) from detector
B LLaVA-style (conversation / detailed description / complex reasoning)
C Evol-Instruct (seed atypical Q →… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/imagenet-variations-synth-pilot-v3.imagenet-variations-synth-pilot-v4
imagenet-variations-synth-pilot-v4
Synthetic multimodal instruction data. Prompts come from laion/imagenet_variations;
images are generated with FLUX.1-schnell, encoded to 32 SEED-2 tokens per image,
grounded with Florence-2, and turned into USER / ASSISTANT records.
This is v4, rebuilt from v3.1 to cover the review feedback of 2026-08-15.
What is new vs v3.1
Axis
v3.1
v4
Reasoning
none
<think> traces on ~50% of records
Prompts per image
1
1-4 turns… See the full description on the dataset page: https://huggingface.co/datasets/EmpathicRobotics/imagenet-variations-synth-pilot-v4.ASCIIfy_ImageNet_Validnano-imagenet-c
Nano ImageNet-C (Severity 5)
This is a randomly sampled subset of the ImageNet-C dataset, containing 5,000 images exclusively from corruption severity level 5. It is designed for efficient testing and validation of model robustness.
这是一个从 ImageNet-C 数据集中随机抽样的子集，包含 5000 张仅来自损坏等级为 5 的图像。它旨在用于高效地测试和验证模型的鲁棒性。
How to Generate / 如何生成
This dataset was generated using the create_nano_dataset.py script included in this repository. To ensure reproducibility, the following… See the full description on the dataset page: https://huggingface.co/datasets/niuniandaji/nano-imagenet-c.ImageNet-DCAE-Tokenizedimagenet-1kImageNet3D-category-captions
ImageNet3D (Category Captions)
This is part of the data release of ImageNet3D. Please refer to our GitHub repo or the huggingface page for more details.
imagenet_variationsgenerated-imagenette
Generated Imagenette Dataset
Description
This repository contains the dataset used for the generative-data-augmentation project. The dataset is organized as follows:
Dataset Structure
analysis/: This directory contains analysis related to the dataset.
metadata/: This directory contains the list of file path used for the Synthetic (Noisy) and Synthetic (Clean) datasets.
synthetic/: This directory contains the image files. Each folder represents a class.… See the full description on the dataset page: https://huggingface.co/datasets/czl/generated-imagenette.