datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
trivia_qa
Dataset Card for "trivia_qa"
Dataset Summary
TriviaqQA is a reading comprehension dataset containing over 650K
question-answer-evidence triples. TriviaqQA includes 95K question-answer
pairs authored by trivia enthusiasts and independently gathered evidence
documents, six per question on average, that provide high quality distant
supervision for answering the questions.
Supported Tasks and Leaderboards
More Information Needed
Languages… See the full description on the dataset page: https://huggingface.co/datasets/lorenzofalappa/trivia_qa.grade-school-math-instructions-Malagasy
Overview
This dataset is a Malagasy adaptation of grade-school-math-instructions.
It consists of arithmetic word problems converted into instruction-answer pairs in Malagasy.
Each entry contains a math problem presented as an instruction, optional contextual input,
and a detailed step-by-step solution in Malagasy.
The dataset is particularly useful for training and evaluating models on arithmetic reasoning and instruction-following tasks in Malagasy, a low-resource language.… See the full description on the dataset page: https://huggingface.co/datasets/Lo-Renz-O/grade-school-math-instructions-Malagasy.LOREA-cyber-eval
LOREA-cyber eval sets
Held-out sets used to benchmark the LOREA-cyber models. Decontaminated 8-gram against the training data,
published so the numbers in the model cards can be reproduced.
These are the sets written for this project. The models are also scored on public benchmarks that aren't
redistributed here: SecQA,
MMLU-Pro,
CyberMetric,
HumanEval.
cyber_mcq (150)
Security knowledge multiple choice across network security, crypto, web/OWASP, malware analysis… See the full description on the dataset page: https://huggingface.co/datasets/MK4-Research/LOREA-cyber-eval.how_lms_answer_one_to_many_factual_queries
One-to-Many Factual Queries Datasets
This is the official dataset used in our EMNLP 2025 paper Promote, Suppress, Iterate: How Language Models Answer One-to-Many Factual Queries.
The dataset includes six subsets named {dataset_name}_template_{i}, where dataset_name is country_cities, artist_songs, or actor_movies, and each dataset has three prompt templates (i = 1, 2, 3).
The {model_name}_step_{i} split in each subset contains the data used for analyzing model_name's behavior at… See the full description on the dataset page: https://huggingface.co/datasets/LorenaYannnnn/how_lms_answer_one_to_many_factual_queries.bannerlord-lore-dataset
Bannerlord Lore Dataset
Comprehensive lore dataset for Mount & Blade II: Bannerlord - a medieval action RPG by TaleWorlds Entertainment.
Dataset Description
This dataset contains structured lore information extracted from the game, including:
Categories
Category
Description
Languages
Heroes
NPCs, lords, companions
EN, RU, TR
Kingdoms
Major factions (Empire, Battania, etc.)
EN, RU, TR
Settlements
Cities, castles, villages
EN, RU, TR
Cultures… See the full description on the dataset page: https://huggingface.co/datasets/TSEOsiris/bannerlord-lore-dataset.alpaca-gpt4-Malagasy
Overview
This dataset is a Malagasy adaptation of the Alpaca-GPT4 instruction-following dataset.It contains instruction-response pairs translated or adapted into Malagasy, designed for fine-tuning instruction-following language models. Each entry includes an instruction, optional input context, and a reference response generated by GPT-4 and adapted to Malagasy using Gemini 2.5 for the translation.
The dataset enables training and evaluating LLMs on instruction understanding… See the full description on the dataset page: https://huggingface.co/datasets/Lo-Renz-O/alpaca-gpt4-Malagasy.