datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
STARK_10k
STARK: Spatial-Temporal reAsoning benchmaRK
STARK is a comprehensive benchmark designed to systematically evaluate large language models (LLMs) and large reasoning models (LRMs) on spatial-temporal reasoning tasks, particularly for applications in cyber-physical systems (CPS) such as robotics, autonomous vehicles, and smart city infrastructure.
Dataset Summary
Hierarchical Benchmark: Tasks are structured across three levels of reasoning complexity:
State Estimation:… See the full description on the dataset page: https://huggingface.co/datasets/prquan/STARK_10k.stark
STaRK
Website | Github | Paper
STaRK is a large-scale semi-structure retrieval benchmark on Textual and Relational Knowledge Bases
Downstream Task
Retrieval systems driven by LLMs are tasked with extracting relevant answers from a knowledge base in response to user queries. Each knowledge base is semi-structured, featuring large-scale relational data among entities and comprehensive textual information for each entity. We have constructed three knowledge bases: Amazon SKB… See the full description on the dataset page: https://huggingface.co/datasets/snap-stanford/stark.STARK_1k
Benchmarking Spatiotemporal Reasoning in Large Language Models: Capabilities and Challenges
Dataset for our paper: Benchmarking Spatiotemporal Reasoning in Large Language Models: Capabilities and Challenges
Contact Information
If you have any questions or feedback, feel free to reach out:
Name: Pengrui Quan
Email: prquan@ucla.edu
License
Copyright (c) 2025, UCLA Networked and Embedded Systems Laboratory (NESL)
All rights reserved.
Redistribution and use in… See the full description on the dataset page: https://huggingface.co/datasets/prquan/STARK_1k.Asclepius-Synthetic-Clinical-Notes
Asclepius: Synthetic Clincal Notes & Instruction Dataset
Dataset Summary
This dataset is official dataset for Asclepius (arxiv)
This dataset is composed with Clinical Note - Question - Answer format to build a clinical LLMs.
We first synthesized synthetic notes from PMC-Patients case reports with GPT-3.5
Then, we generate instruction-answer pairs for 157k synthetic discharge summaries
Supported Tasks
This dataset covers below 8 tasks
Named Entity… See the full description on the dataset page: https://huggingface.co/datasets/starmpcc/Asclepius-Synthetic-Clinical-Notes.startup-Investments-analysis
📊 StartUp Investments EDA
1. Background & Objectives
This project explores a comprehensive dataset of startup investments (sourced from Crunchbase) to uncover the primary factors that predict a startup's survival and trajectory in a competitive market.
Through this Exploratory Data Analysis (EDA), we analyze historical funding data, investment rounds, and market categories to determine which variables drive specific company outcomes - namely, whether a business… See the full description on the dataset page: https://huggingface.co/datasets/lia-prop13/startup-Investments-analysis.starcop_allbands_mini
MINI version of the STARCOP dataset
For full details please refer to https://huggingface.co/datasets/previtus/STARCOP_allbands_Train1
STARCOP_allbands_Train1
STARCOP dataset
STARCOP dataset: Semantic Segmentation of Methane Plumes with Hyperspectral Machine Learning Models 🌈🛰️Authors: Vít Růžička, Gonzalo Mateo-Garcia, Luis Gómez-Chova, Anna Vaughan, Luis Guanter and Andrew Markham
Fast data preview in: dataset_exploration.ipynb Main repository: github/spaceml-org/STARCOP
Task:
Methane is the second most important greenhouse gas contributor to climate change; at the same time its reduction has been denoted as one of the… See the full description on the dataset page: https://huggingface.co/datasets/previtus/STARCOP_allbands_Train1.STaR
Systematic, Multipath, disjunctive Spatio-Temporal Reasoning (STaR) benchmark
Main idea
This is part of the paper that was published in ICLR 2025 where further details can be found.
The main idea is to expand the concept of binary relational composition to not just be atomic and require that a model reason over multiple
paths instead of a single one in contrast with previous art.
This can be seen from the example below contrasting a popular previous benchmark.… See the full description on the dataset page: https://huggingface.co/datasets/erg0dic/STaR.STARCOP_allbands_Eval
STARCOP dataset
STARCOP dataset: Semantic Segmentation of Methane Plumes with Hyperspectral Machine Learning Models 🌈🛰️Authors: Vít Růžička, Gonzalo Mateo-Garcia, Luis Gómez-Chova, Anna Vaughan, Luis Guanter and Andrew Markham
Fast data preview in: dataset_exploration.ipynb Main repository: github/spaceml-org/STARCOP
Please refer to the main dataset readme file on: https://huggingface.co/datasets/previtus/STARCOP_allbands_Train1
Traffic-sign-detection-VietNam
Vietnam Traffic Sign Detection Dataset
This repository contains the dataset for detecting road traffic signs in Vietnam using the state-of-the-art YOLO object detection model.
📂 Repository Structure
The dataset is structured in the standard YOLO format, containing images and corresponding annotations divided into training, validation, and testing sets.
├── classid.xlsx # Excel file mapping class IDs to names
├── dataset/
│ ├── train/ #… See the full description on the dataset page: https://huggingface.co/datasets/star092304/Traffic-sign-detection-VietNam.typhoon-intensity-classification
Typhoon - Image Classification Dataset
This dataset comes from PTIT AI Challenge and is organized for a multi-class image classification task focusing on tropical cyclone (typhoon) intensity estimation.
Dataset Structure
The directory structure is organized as follows:
train/
├── images/
│ ├── image1.jpg
│ └── ...
└── annotations.csv (only present in the train folder)
The public_test and private_test sets are used to evaluate and score the… See the full description on the dataset page: https://huggingface.co/datasets/star092304/typhoon-intensity-classification.ViSignLanguage-Video
AI Challenge CV Dataset Description
This dataset comes from https://aichallenge.ptit.edu.vn/ and is organized for a multi-class video classification task focusing on Vietnamese Sign Language.
🛠 Dataset Viewer Configuration
This dataset supports Hugging Face Dataset Viewer. The dataset structure and video paths are mapped via dataset_metadata.csv.
Expected CSV Structure
To ensure the Dataset Viewer renders correctly, your dataset_metadata.csv… See the full description on the dataset page: https://huggingface.co/datasets/star092304/ViSignLanguage-Video.cas9_cas12_starter_dataThis dataset contains 1,206 Cas9 and Cas12 sequences from Madugula et al.
In starter_cas_data.csv, the entries are:
sequence: (string) the amino acid sequence of the Cas protein
sequence_len: (int) length of the sequence in number of amino acids
cas: (string) Cas 9 or Cas12
CAS_TYPE: (int) 0 if Cas9, 1 if Cas12
startups
VentureMatch — Synthetic Startups Dataset (10,000 rows)
This repository contains a synthetic dataset of 10,000 startups, generated using an open-source LLM and enriched with financial variables to enable realistic economic and venture-capital analysis.
The same notebook includes both the synthetic data generation pipeline and the exploratory data analysis (EDA).
Dataset Overview
Each row represents a synthetic startup with both textual and numerical attributes.… See the full description on the dataset page: https://huggingface.co/datasets/Yoav-omer/startups.internship-starter
FlyRank Internship — Starter Dataset (Anonymized)
The public, safe starting point for the FlyRank Applied Search Intelligence ML internship.
30,000 anonymized content-performance rows across 32 pseudonymized clients (53 columns).
Public-safe: hashed content_id / client_id + numeric/categorical metrics only — no titles, URLs, keywords, domains, or client names.
What it's for
Week 1–2 quick wins and the ready-now capstone lanes (ranking-signal analysis, lifecycle /… See the full description on the dataset page: https://huggingface.co/datasets/FlyRank/internship-starter.shopee-reviews-tl-stars
Dataset Card for Dataset Name
Dataset Summary
This dataset card aims to be a base template for new datasets. It has been generated using this raw template.
Supported Tasks and Leaderboards
[More Information Needed]
Languages
Tagalog (TL)
Dataset Structure
Data Instances
A typical data point, comprises of a text and the corresponding label.
An example from the YelpReviewFull test set looks as follows:
{
'label': 2… See the full description on the dataset page: https://huggingface.co/datasets/scaredmeow/shopee-reviews-tl-stars.telegram-public-channels-2026-W36
Telegram public channels: a 5,177-channel snapshot with topics and a recommendation graph
A single snapshot of 5,177 public Telegram channels, measured on 31 August 2026, together with the
recommendation graph Telegram itself exposes between them.
Public Telegram data is hard to get in tabular form. The libraries that read it need a phone number
and a user session, and the two Telegram datasets that rank on Kaggle today are both from 2021.
This is a current measurement… See the full description on the dataset page: https://huggingface.co/datasets/starnikovoleg/telegram-public-channels-2026-W36.stargo-embeddings
stargo-embeddings
Dataset repository containing STAR-GO related embedding assets. The metadata.csv is loadable via datasets.load_dataset, while large binaries (e.g. .h5, .npy) are stored as downloadable files.
How to use
Load the metadata table:
from datasets import load_dataset
ds = load_dataset("<your-org-or-username>/<your-dataset-repo>")
print(ds)
Download the large binary assets referenced in the table with hf_hub_download.
MSR_data_cleaned
MSR Data Cleaned - C/C++ Code Vulnerability Dataset
📌 Dataset Description
A curated collection of C/C++ code vulnerabilities paired with:
CVE details (scores, classifications, exploit status)
Code changes (commit messages, added/deleted lines)
File-level and function-level diffs
🔍 Sample Data Structure from original file
+---------------+-----------------+----------------------+---------------------------+
| CVE ID | Attack Origin | Publish Date… See the full description on the dataset page: https://huggingface.co/datasets/starsofchance/MSR_data_cleaned.Dense-Information-Science-Physics-Dataset
Dense Information With Multiple Fine-tuned Variations
This dataaset has multiple for each input to learn how to express the same answer in different ways
Dataset Structure
The dataset contains two columns:
Column
Description
input
A science or quantum-physics question
output
A conversational answer to the question
Example:
{
"input": "What is quantum entanglement?",
"output": "Quantum entanglement is when two quantum systems share one… See the full description on the dataset page: https://huggingface.co/datasets/StarpowerTechnology/Dense-Information-Science-Physics-Dataset.StartYourOwnGoldMine-Sampling-Series-Datasetllm-cold-start-benchmark
LLM Container Cold-Start Benchmark
Measurements of how long it takes to bring a language model from cold storage to
a state where it can serve its first token, across 25 open-weight
models spanning 17 architecture families and
100.9 GiB of checkpoints, on a single NVIDIA T4.
Cold start is the latency a serverless or scale-to-zero inference platform pays
when it has no warm replica. It decomposes into weight transfer from storage,
deserialization into host memory, transfer to the… See the full description on the dataset page: https://huggingface.co/datasets/ArchCoder/llm-cold-start-benchmark.yc_startupsTry out the YC Chatbot
both-star-5c13b7
both-star-5c13b7
Synthetic products test data: 38 rows in data.csv.
All values are randomly generated fictional examples, not real observations, products, or user activity. Intended only for CSV loading and pipeline tests; not suitable for scientific or business conclusions. Columns are sampled independently and do not model real-world correlations.
Fields
sample_id: random identifier for this generated sample.
row_id: sequential row number starting at 1.… See the full description on the dataset page: https://huggingface.co/datasets/Indigo-Michael/both-star-5c13b7.star-wars-dataset
Dataset Source
This dataset is derived from the Star Wars Scripts Database available on Kaggle.Title: Star Wars Scripts DatabaseCreator: Oscar YezfeijoSource: Kaggle - Star Wars Scripts DatabaseLicense: Open Database (ODbL) for the database, Database Contents under Database Contents License. Check Kaggle Licensing Information for details.
startup-investor-and-accelerator-dataset
Startup Investor & Accelerator Dataset
Overview
The Startup Investor & Accelerator Dataset is a curated collection of startup ecosystem stakeholders, including venture capital firms, angel investors, investment funds, and accelerators. The dataset is designed to help founders, researchers, startup enthusiasts, and ecosystem builders discover potential funding sources and support programs.
Dataset Contents
The dataset contains information about… See the full description on the dataset page: https://huggingface.co/datasets/IqraSAYEDhassan/startup-investor-and-accelerator-dataset.alpaca-star-asciisame as the original alpaca star, this one however encourages to include a mental image.
it will out put a flow chart or ascii image for each prompt
algozee_analysis-of-high-starred-github-repositories
Analysis of High-Starred GitHub Repositories
A comprehensive overview of repository metrics and developer engagement
Dataset Info
Source: Kaggle
Original Size: 0.41 MB
Kaggle Downloads: 14
Files: 1
Files
github_top_repositories.csv
Mirrored from Kaggle
startup-credibility-claim-matrix
Startup Credibility Claim Matrix
This open dataset gives founders a machine-readable way to connect the exact
claim an outside reader sees with its evidence, method, date, state,
limitations and human owner.
Dataset Structure
The template configuration contains one blank starter row for each claim
class:
identity;
quantitative;
comparative;
capability;
outcome.
The fictional-example configuration shows five review states using the
synthetic company Fictional… See the full description on the dataset page: https://huggingface.co/datasets/mheilimo/startup-credibility-claim-matrix.aware-police-5c70c3
aware-police-5c70c3
Synthetic sensors test data: 58 rows in data.csv.
All values are randomly generated fictional examples, not real observations, products, or user activity. Intended only for CSV loading and pipeline tests; not suitable for scientific or business conclusions. Columns are sampled independently and do not model real-world correlations.
Fields
sample_id: random identifier for this generated sample.
row_id: sequential row number starting at 1.… See the full description on the dataset page: https://huggingface.co/datasets/Lunar-Star/aware-police-5c70c3.