A Japanese QA dataset for evaluating Retrieval‑Augmented Generation (RAG), built from JAQKET questions and Wikipedia passages with gold retrieval‑relevance labels, released on HuggingFace and GitHub and scored primarily by nDCG@10.
Description in other languages
Statistics
Statistics as of Nov 16, 2026
Hugging Face downloads (last 30 days)
436
Hugging Face likes
21
Listed
May 13, 2024
Downloads (30-day window) over time
Show as a table
| Month | DL (30d) |
|---|---|
| Dec 2025 | 403 |
| Jan 2026 | 251 |
| Feb 2026 | 277 |
| Mar 2026 | 139 |
| Apr 2026 | 181 |
| May 2026 | 363 |
| Jun 2026 | 248 |
| Jul 2026 | 332 |
| Aug 2026 | 559 |
| Sep 2026 | 321 |
| Oct 2026 | 791 |
| Nov 2026 | 436 |
Downloads over the 30 days before the last snapshot of each month.
Popular in the same category
- Japanese-RAG-Generator-Benchmarkneoai-inc113Japanese RAG Generator Benchmark (J‑RAGBench) supplies a multi‑category QA dataset—covering Integration, Reasoning, Logical, Table, and Abstention—designed to evaluate Japanese RAG generators, built with human effort and GPT‑4.1, and released under CC BY‑SA 4.0.
- JGLUEshunk0313kUpdated JGLUE dataset card and loading script for a Japanese NLP benchmark (created by Yahoo Japan and Waseda University) that covers text classification (MARC‑ja, JCoLA), sentence‑pair classification (JNLI), and QA (JSQuAD, JCommonsenseQA), with releases linked on GitHub and Hugging Face.
- JMTEBsbintuitions3kJMTEB is a Japanese text‑embedding benchmark featuring 5 tasks (clustering, classification, STS, retrieval, reranking) and 28 datasets, offering a one‑line evaluation script and inviting community contributions.
- JamC-QAsbintuitions2kJamC‑QA is a bilingual benchmark of multiple‑choice questions spanning eight Japanese‑culture and knowledge categories, with leaderboard metrics comparing state‑of‑the‑art models.
- JMMMU-ProJMMMU1kJMMMU‑Pro is an image‑based Japanese multimodal benchmark created by generating visual questions with a realistic model and human verification, revealing that current open‑source LMMs perform poorly on it and offering a cost‑effective method for future VQA benchmark development.
- JMedBenchColdog23331kJMedBench is a Japanese biomedical LLM benchmark comprising 20 datasets across five tasks (MCQA, NER, STS, etc.) sourced from MedMCQA, PubMedQA, MMLU, and others, each with its own license, and includes a note that translations may contain biases requiring human review.
Resources in similar research fields
- JEMHopQAsbintuitions297Japanese multi‑hop question‑answering dataset (JEMHopQA) that provides questions, answers and semi‑structured derivation steps linking two Wikipedia articles, with updated derivation formatting and expanded answer sets across multiple versioned releases.Shared: Natural Language Interfaces・Question Answering
- rakuda-questionsyuzuai188Rakuda supplies 40 Japanese questions—open‑ended for history, society, and government, and specific for geography—for benchmarking Japanese AI assistants, comparable to vicuna‑eval, and can be loaded with datasets.load_dataset.Shared: Natural Language Interfaces・Question Answering
ja-vicuna-qa-benchmarkku-nlp34Japanese Vicuna QA BenchmarkGitHubShared: Natural Language Interfaces・Question Answering- JaQuADSkelterLabsInc916JaQuAD is a 2022 Japanese QA dataset of 39,696 SQuAD‑style extractive pairs from Wikipedia, totaling 73.2 MB, that achieves 78.92 % F1 (63.38 % EM) when fine‑tuned with BERT‑Japanese.Shared: Natural Language Interfaces・Question Answering
asdcmegagonlabs25Accommodation Search Dialog Corpus (宿泊施設探索対話コーパス)GitHubShared: Natural Language Interfaces・Information Retrieval- reranker-scoreshpprc288Provides a Japanese search/QA dataset with per‑query scores computed by five multilingual/Japanese rerankers (e.g., BAAI/bge‑reranker‑v2‑m3, Alibaba‑NLP/gte‑multilingual‑reranker‑base), including average scores for roughly 200 positive and negative example documents per query.Shared: Information Retrieval