datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
answercarefully-dpo-ja-2026
AnswerCarefully-derived Japanese DPO data for LLM safety
本データセットは、llm-jp/AnswerCarefullyを参照して作成した日本語LLMの安全応答をDPOで学習するためのpreference datasetです。
利用条件
本データセットには、llm-jp/AnswerCarefullyと同じ利用規約を適用します。
利用者は、llm-jp/AnswerCarefullyと本データセットの両方で利用規約に同意する必要があります。
データ
train: 417件
validation: 44件
各行には次のフィールドが含まれます。
id: 本リリース内だけで使用するID
prompt: 元質問の意味と危険性を変えずに言い換えた質問
chosen: DPOで望ましい応答として扱う回答
rejected: DPOで望ましくない応答として扱う回答
category, harm_type, risk_area… See the full description on the dataset page: https://huggingface.co/datasets/ekunish/answercarefully-dpo-ja-2026.text-sft-questions-answers-only
text-sft: Questions and Answers
This dataset consists of question-and-answer pairs generated from short excerpts drawn from Wikipedia, Cosmopedia, and FineWeb-Edu. It is an adapted version of agentlans/text-sft.
Overview
The dataset provides compact examples of English question-and-answer relationships that can help models learn linguistic patterns, syntactic structures, and semantic associations between questions and their corresponding answers.
Intended Use… See the full description on the dataset page: https://huggingface.co/datasets/agentlans/text-sft-questions-answers-only.reddit_question_best_answersQuestion & question body together with the best answers to that question from Reddit.
The score for the question / answer is the upvote count (i.e. positive-negative upvotes).
Only questions / answers that have these properties were extracted:
min_score = 3
min_title_len = 20
min_body_len = 100
cybersecurity_full_question_answersgpqa-metadata-blind-answerscugnizz-v22-tool-answer
scugnizz-v22-tool-answer
Synthetic tool-loop data teaching grounded answers and correct refusal on empty/error tool results.
Format: Hermes/OpenAI-style messages plus tools.
do-not-answer-scenarios
Do-Not-Answer Scenarios
Refusal-evaluation scenarios derived from the Do-Not-Answer dataset, serialized as
giskard Scenario objects (one JSON object per line).
Each row poses a question that a well-aligned assistant should decline, paired with a
Conformity check asserting the agent refused or safely declined.
Format
Every line is a serialized giskard.checks.Scenario:
name — "Do not answer #<id>"
steps[].interacts[].inputs — the question to send to the agent under… See the full description on the dataset page: https://huggingface.co/datasets/giskardai/do-not-answer-scenarios.quora-question-answer-datasetQuora Question Answer Dataset (Quora-QuAD) contains 56,402 question-answer pairs scraped from Quora.
Usage:
For instructions on fine-tuning a model (Flan-T5) with this dataset, please check out the article: https://www.toughdata.net/blog/post/finetune-flan-t5-question-answer-quora-dataset
answersumm
Dataset Card for answersumm
Dataset Summary
The AnswerSumm dataset is an English-language dataset of questions and answers collected from a StackExchange data dump. The dataset was created to support the task of query-focused answer summarization with an emphasis on multi-perspective answers.
The dataset consists of over 4200 such question-answer threads annotated by professional linguists and includes over 8700 summaries. We decompose the task into several annotation… See the full description on the dataset page: https://huggingface.co/datasets/alexfabbri/answersumm.gsm8k_only_answerThe data is exactly like the original GSM8k (https://huggingface.co/datasets/gsm8k ), but with the label consisting of the correct answer(one number) only.
@misc{krishna2024gsmansweronly,
title={GSM8k (Answer only)},
author={Satyapriya Krishna},
year={2023},
url={skrishna/gsm8k_only_answer},
}
minecraft-question-answer-700k
minecraft-question-answer-700k
Introducing the largest synthetic Minecraft Q&A dataset, covering every topic, game mechanic, item and craft in Minecraft. The dataset was generated by extracting over 18,000 Minecraft wiki pages, and using glaive.ai's synthetic data generation pipeline.
about the dataset
rows - 694,814
tokens - 47,133,624
source - https://minecraft.wiki/
Hit me up on twitter if you see a bug or need a synthetic dataset for your company:… See the full description on the dataset page: https://huggingface.co/datasets/naklecha/minecraft-question-answer-700k.CFQA_Chinese_Finance_Question_Answering
Citation
For the complete project, please check Here
If you use CFQA in your research, experiments, benchmarks, or publications, please cite the accompanying paper:
@inproceedings{zhu2026cfqa,
title = {CFQA: A Chinese Financial Question Answering Benchmark From Corporate Annual Reports},
author = {Tianning Zhu and Mo Liu and Murathan Kurfali},
booktitle = {Proceedings of The 7th Financial Narrative Processing Workshop (FNP 2026)},
year = {2026},
address =… See the full description on the dataset page: https://huggingface.co/datasets/ZackZhu00/CFQA_Chinese_Finance_Question_Answering.question-answering-state-of-the-unionnatural-questions-slim-short-answer
Natural Questions Slim Short Answer
This is a slim, flattened derived version of
google-research-datasets/natural_questions
for short-answer question answering experiments.
The conversion keeps examples with extractable short answers and removes the
original document HTML, token-level document spans, long answer candidates, and
yes/no-only examples. Each record is a simple question-answer pair. It is
intended for lightweight QA prompting and evaluation, not as a full replacement
for… See the full description on the dataset page: https://huggingface.co/datasets/BOB12311/natural-questions-slim-short-answer.kor_ethical_question_answerIMO-AnswerBench-Verified
IMO AnswerBench Verified
IMO AnswerBench Verified is a human-expert-verified derivative of OpenEvals/IMO-AnswerBench, originally curated by the Google DeepMind Superhuman Reasoning team. Every record in the 400-problem benchmark was reviewed individually. The review identified and corrected 13 records while preserving the benchmark's balanced coverage of four major mathematical areas.
Dataset summary
Total records: 400
Verification method: record-by-record human… See the full description on the dataset page: https://huggingface.co/datasets/dots-studio/IMO-AnswerBench-Verified.data_gpt54_only_answer_lossreddit_question_best_answers_langschaii-hindi-and-tamil-question-answeringflutter-questions-answerspsychology-question-answerA JSON formatted dataset comprising 197,180 question and answer pairs covering a wide range of topics encountered in a Bachelor level psychology course. I have included a broad range of question types, topics, and answer styles.
The dataset was created using personal notes and several LLMs (such as GPT4) and manually assessed for veracity and completeness of response. Despite this, the size of the dataset prohibits me from ensuring every single answer is 100% accurate and up-to-date. As such… See the full description on the dataset page: https://huggingface.co/datasets/BoltMonkey/psychology-question-answer.answer-equivalence-dataset
Answer Equivalence Dataset
This dataset is introduced and described in Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation.
Source
This is a repost. The original dataset repository can be found here.
Data splits and sizes
AE Split
# AE Examples
# Ratings
Train
9,090
9,090
Dev
2,734
4,446
Test
5,831
9,724
Total
17,655
23,260
Split by system
# AE Examples
# Ratings
BiDAF dev predictions
5622… See the full description on the dataset page: https://huggingface.co/datasets/kortukov/answer-equivalence-dataset.bodo-legal-question-answering-ai4bharat
Bodo Legal Question Answering Dataset
Overview
This dataset is a Bodo-language legal Question Answering (QA) resource
created for research in low-resource Natural Language Processing (NLP)
and legal language processing.
The supplied source files contain legal judgment contexts together with
multiple questions and answers. For Hugging Face compatibility and
question-answering model training, each question-answer pair has been
flattened into a separate JSONL example… See the full description on the dataset page: https://huggingface.co/datasets/Mwnthai/bodo-legal-question-answering-ai4bharat.question-answering-paul-grahamIELTs-Speaking-answer
Overview
This dataset consists of 2 json files named 'ielts_new.json' and 'ielts_old.json', which contain ielts questions and its corresponding answers for part 1 and part 2.
'ielts_new.json': new IELTs topics for 2024 September-December.
'ielts_old.json': remained IELTs topics for 2024 September-December.
Quality
Since the dataset is analysed and generated by ChatGPT based on my own pdf file, the answer may be incomplete(only part of the sentence is extracted, leading to… See the full description on the dataset page: https://huggingface.co/datasets/qwertyuiopasdfg/IELTs-Speaking-answer.open_parallel_think_cot_update_wo_answer
open_parallel_think_cot_update_wo_answer
This dataset is derived from haowu89/open_parallel_think_cot_update.
Transformation applied:
For every example, for every string item inside context, remove the final sentence.
The intent is to strip the trailing answer-bearing sentence while keeping the earlier reasoning trajectory.
Generated on 2026-04-15.
2026-09-16-da-7-answer-only-mix
DA supervision answer; all 752 DA and 9284 identical replay rows
field
value
experiment
DA supervision answer; all 752 DA and 9284 identical replay rows
date_generated
2026-09-16
constitution
constitutions/claude_distilled_09_principles/constitution.md
source_repo
https://github.com/Matthew-Bozoukov/teaching_claude_why_replication.git @ 4648153af4b834b70bd2e5374f639aaad219c83c
models
Tokenizer Qwen/Qwen3.6-27B@6a9e13bd6fc8f0983b9b99948120bc37f49c13e9; replay… See the full description on the dataset page: https://huggingface.co/datasets/dougalldeepmind/2026-09-16-da-7-answer-only-mix.Magpie-Tanuki-Qwen2.5-72B-Answered
Magpie-Tanuki-Qwen2.5-72B-Answered
Aratako/Magpie-Tanuki-8B-annotated-96kからinput_qualityがexcellentのものを抽出し、それに対してQwen/Qwen2.5-72B-Instructで回答の再生成を行ったデータセットです。
ライセンス
基本的にはApache 2.0に準じますが、Qwen Licenseの影響を受けるため、このデータセットを使ってモデルを学習する際はこのライセンスの制約に従ってください。
hico-det-llava-v1.6-13b-answers
HICO-DET LLaVA-1.6-13B action answers
Per-image action lists for every HICO-DET image (38,118 train / 9,658 test),
produced by LLaVA-1.6 (vicuna-13B) prompted with the 117 HICO-DET verb
names and asked to list at most 7 valid actions visible in the picture.
These are the text-side VLM answers consumed by UMI-HOI (Unified
Multimodal Interaction HOI detection) at training and test time.
No images are included; obtain HICO-DET separately and join on image.
Files… See the full description on the dataset page: https://huggingface.co/datasets/Pikaqiu0114/hico-det-llava-v1.6-13b-answers.do-not-answer-jaThis dataset was created by automatically translating "do-not-answer" into Japanese.This dataset is licensed under CC-BY-NC-SA-4.0
do-not-answer-jahttps://github.com/kunishou/do-not-answer-jado-not-answerhttps://github.com/Libr-AI/do-not-answer