datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Language-Grounded_Sparse_Encoder_Training
Language-Grounded Sparse Encoder (LanSE) — Training Data
This repository hosts the AI-generated images and human annotation datasets accompanying the paper:
Human-like Content Analysis for Generative AI with Language-Grounded Sparse Encoders
Yiming Tang, Arash Lagzian, Srinivas Anumasa, Qiran Zou, Yingtao Zhu, Ye Zhang, Trang Nguyen, Yih-Chung Tham, Ehsan Adeli, Ching-Yu Cheng, Yilun Du, Dianbo Liu
National University of Singapore · Tsinghua University · Stanford University ·… See the full description on the dataset page: https://huggingface.co/datasets/DesmondYMTang2024/Language-Grounded_Sparse_Encoder_Training.OpenSDI_trainThis repository contains the OpenSDI training dataset, presented in the paper OpenSDI: Spotting Diffusion-Generated Images in the Open World.
Code: https://github.com/iamwangyabin/OpenSDI
BioTrove-Train
BioTrove: A Large Curated Image Dataset Enabling AI for Biodiversity
Description
See the BioTrove dataset card on HuggingFace to access the main BioTrove dataset (161.9M)
BioTrove comprises well-processed metadata with full taxa information and URLs pointing to image files. The metadata can be used to filter specific categories, visualize data distribution, and manage imbalance effectively. We provide a collection of software… See the full description on the dataset page: https://huggingface.co/datasets/BGLab/BioTrove-Train.rlbenchfail_train_dataset
Guardian: RLBench-Fail Dataset
This dataset is part of the Guardian project: Detecting Robotic Planning and Execution Errors with Vision-Language Models. It contains annotated robotic manipulation failure data generated in the RLBench simulator for training and evaluating Vision-Language Models (VLMs) on failure detection tasks.
Failures are produced by an automated pipeline that procedurally perturbs successful scripted trajectories in simulation, generating diverse planning… See the full description on the dataset page: https://huggingface.co/datasets/paulpacaud/rlbenchfail_train_dataset.bdv2fail_train_dataset
Guardian: BridgeDataV2-Fail Dataset
This dataset is part of the Guardian project: Detecting Robotic Planning and Execution Errors with Vision-Language Models. It contains annotated robotic manipulation failure data derived from the BridgeDataV2 real-robot dataset for training and evaluating Vision-Language Models (VLMs) on failure detection tasks.
Failures are produced by an automated pipeline that perturbs successful real-robot trajectories offline (without re-executing actions)… See the full description on the dataset page: https://huggingface.co/datasets/paulpacaud/bdv2fail_train_dataset.ur5fail_train_dataset
Guardian Failure Detection Dataset
This dataset is part of the Guardian project: Detecting Robotic Planning and Execution Errors with Vision-Language Models. It contains annotated robotic manipulation failure data for training and evaluating Vision-Language Models (VLMs) on failure detection tasks.
Guardian introduces an automated failure generation approach that procedurally perturbs successful robot trajectories to produce diverse planning failures and execution failures, each… See the full description on the dataset page: https://huggingface.co/datasets/paulpacaud/ur5fail_train_dataset.oven-grpo-training-data
OVEN GRPO Training Data
Training data behind my MSc thesis GRPO experiments on OVEN open-domain visual
entity recognition with Qwen3-VL-4B. This is the companion to the published
GRPO adapters: the parquets,
taxonomy oracle, and self-aggregation traces they were trained on.
Layout
path
contents
taxonomy/chains_labels.jsonl
Wikidata P279 (subclass-of) chains per OVEN entity: the label hierarchy used by every hierarchical metric… See the full description on the dataset page: https://huggingface.co/datasets/jucamohedano/oven-grpo-training-data.training
87cc2s/training
Numerosity training data, organized by task then by source:
counting-training/
synthetic/ <- abstract shapes generator
real-world/ <- real photographs, ~19 public counting/detection datasets
ans-training/
synthetic/ <- abstract shapes generator
real-world/ <- same-category pairs of real photographs
Each *-training/<domain>/ folder has its own dataset_card.md (schema +
generation/curation details), manifest.json (summary stats), and… See the full description on the dataset page: https://huggingface.co/datasets/87cc2s/training.lanternfly_swatter_training
Spotted Lanternfly Classification Dataset
Dataset Description
This dataset contains images for binary classification of spotted lanternflies (Lycorma delicatula), an invasive species causing significant damage to agriculture and ecosystems in the United States. The dataset is designed to train machine learning models to identify dead or squashed lanternflies from photographs, supporting community-driven environmental monitoring and pest management efforts.
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/rlogh/lanternfly_swatter_training.skin-lesion-trainOne-full
Skin Lesion Dataset — trainOne Full
All 7 HAM10000 classes including Melanocytic Nevi (nv).
Source: ISIC 2018 Challenge Task 3.
Code
Description
Count
nv
Melanocytic Nevi (benign)
5000
mel
Melanoma (deadly)
1113
bkl
Benign Keratosis
1099
bcc
Basal Cell Carcinoma (cancer)
514
akiec
Actinic Keratosis (monitor)
327
vasc
Vascular Lesions
142
df
Dermatofibroma
115
Total: 8,310 images
Load:
from datasets import load_dataset
ds =… See the full description on the dataset page: https://huggingface.co/datasets/zujiguts/skin-lesion-trainOne-full.skin-lesion-trainOne
Skin Lesion Dataset — trainOne
Subset of the ISIC 2018 Challenge (HAM10000) dataset.
Melanocytic Nevi (nv) excluded — handled as a separate class.
Code
Full Name
Note
mel
Melanoma
Deadly
bkl
Benign Keratosis
bcc
Basal Cell Carcinoma
Cancer
akiec
Actinic Keratosis
Monitor
vasc
Vascular Lesions
df
Dermatofibroma
Source: ISIC 2018 Challenge Task 3 — https://challenge.isic-archive.com/data/#2018
Load this dataset:
from datasets import load_dataset
ds =… See the full description on the dataset page: https://huggingface.co/datasets/zujiguts/skin-lesion-trainOne.
