Datasets:
Add dataset card
Browse files
README.md
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-sa-3.0
|
| 3 |
+
task_categories:
|
| 4 |
+
- question-answering
|
| 5 |
+
- text-retrieval
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
tags:
|
| 9 |
+
- wikipedia
|
| 10 |
+
- agentic-search
|
| 11 |
+
- retrieval
|
| 12 |
+
- offline-index
|
| 13 |
+
pretty_name: SimpleWikiSearch
|
| 14 |
+
size_categories:
|
| 15 |
+
- 10M<n<100M
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
# SimpleWikiSearch Indexes
|
| 19 |
+
|
| 20 |
+
Offline Wikipedia indexes for [SimpleWikiSearch](https://github.com/JimXiongGM/simple_wiki_search) (arXiv:[2607.26070](https://arxiv.org/abs/2607.26070)).
|
| 21 |
+
|
| 22 |
+
These files are **binary search indexes** (not row-oriented tables). Download and extract them locally; they are not intended for `load_dataset()`.
|
| 23 |
+
|
| 24 |
+
Paper page: https://huggingface.co/papers/2607.26070
|
| 25 |
+
|
| 26 |
+
## Files
|
| 27 |
+
|
| 28 |
+
| Archive | Size (approx.) | Contents |
|
| 29 |
+
|---------|----------------:|----------|
|
| 30 |
+
| `enwiki-20260601-tantivy-chunk-1536.tar.zst` | ~22 GB | Tantivy keyword index + chunk text store |
|
| 31 |
+
| `enwiki-20260601-faiss-hnsw-Qwen3-Embedding-0.6B.tar.zst` | ~19 GB | FAISS HNSW dense index (`Qwen3-Embedding-0.6B`, dim=1024, fp16) |
|
| 32 |
+
|
| 33 |
+
Corpus snapshot: `enwiki-20260601` — **7,189,602** articles / **10,837,506** section-aware chunks (target 1536 tokens).
|
| 34 |
+
|
| 35 |
+
## Extract
|
| 36 |
+
|
| 37 |
+
```bash
|
| 38 |
+
tar -I zstd -xf enwiki-20260601-faiss-hnsw-Qwen3-Embedding-0.6B.tar.zst
|
| 39 |
+
tar -I zstd -xf enwiki-20260601-tantivy-chunk-1536.tar.zst
|
| 40 |
+
```
|
| 41 |
+
|
| 42 |
+
Expected layout:
|
| 43 |
+
|
| 44 |
+
```text
|
| 45 |
+
data/database/wikipedia-index/enwiki-20260601/
|
| 46 |
+
faiss/hnsw-Qwen3-Embedding-0.6B-chunk-1536-fp16/
|
| 47 |
+
index.faiss
|
| 48 |
+
ids.pkl
|
| 49 |
+
tantivy/chunk-1536/
|
| 50 |
+
chunks/
|
| 51 |
+
titles/
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
Full setup (embedding server, MCP tools, QA agent): see [deploy.md](https://github.com/JimXiongGM/simple_wiki_search/blob/main/deploy.md) in the GitHub repo.
|
| 55 |
+
|
| 56 |
+
## License
|
| 57 |
+
|
| 58 |
+
Wikipedia text is available under [CC BY-SA 3.0](https://creativecommons.org/licenses/by-sa/3.0/). Indexes derived from that dump inherit the same license obligations.
|
| 59 |
+
|
| 60 |
+
## Citation
|
| 61 |
+
|
| 62 |
+
```bibtex
|
| 63 |
+
@misc{xiong2026simplewikisearchcleanofflinewikipedia,
|
| 64 |
+
title={SimpleWikiSearch: A Clean Offline Wikipedia Environment for Agentic Search},
|
| 65 |
+
author={Guanming Xiong and Penghui Zhang},
|
| 66 |
+
year={2026},
|
| 67 |
+
eprint={2607.26070},
|
| 68 |
+
archivePrefix={arXiv},
|
| 69 |
+
primaryClass={cs.IR},
|
| 70 |
+
url={https://arxiv.org/abs/2607.26070},
|
| 71 |
+
}
|
| 72 |
+
```
|