IdeaShift-X: data and card

README.md

@@ -0,0 +1,109 @@
+---
+license: other
+license_name: mixed-see-card
+language:
+- am
+- ar
+- bn
+- de
+- es
+- fa
+- fr
+- hi
+- id
+- ja
+- km
+- ml
+- mr
+- ne
+- ru
+- si
+- sw
+- ta
+- th
+- uk
+- ur
+- vi
+- yo
+- zh
+tags:
+- ai-text-detection
+- idea-provenance
+- multilingual
+pretty_name: IdeaShift-X
+size_categories:
+- 10K<n<100K
+extra_gated_prompt: "Access is granted individually. Please say who you are and what you intend to use the data for."
+configs:
+- config_name: default
+  data_files:
+  - split: test
+    path: data/ideashift_x.parquet
+---
+
+# IdeaShift-X
+
+IdeaShift-X tests idea-provenance detection outside English. It holds about 10,000 human-written web pages in 24
+languages from [FineWeb2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2). Each page has two AI versions in
+the same language, written by Gemini 3.7 Flash:
+
+- **Level 0:** written from a brief naming only the document type, so the ideas are the AI's.
+- **Level 5:** written from the page's full role-labelled outline with an instruction to add nothing, so the ideas
+  are the page's and the prose is the AI's.
+
+A detector should not flag the human pages. It should flag level 0 for AI ideas. An idea detector should not flag
+level 5; a prose detector should.
+
+## Construction
+
+The dataset merges two builds.
+
+- **9-language build:** Arabic, Chinese, German, Indonesian, Japanese, Russian, Spanish, Tamil, Urdu. About 610-740
+  pages per language, sampled from crawls dated before 2022.
+- **24-language build:** adds 15 languages at about 250 pages each. This build was **not** date-filtered. Its pages
+  come from FineWeb2 crawls up to 2024, and 321 of them were crawled in 2022-2024. Use `seed_crawl_date` to filter
+  them.
+
+Pages were format-classified with WebOrganizer's annotation prompt on an LLM. Only pages in the eight long-form
+formats IdeaLens was trained on were kept, some of them force-fitted to the closest format. Outlines are always
+extracted in English, whatever the page's language.
+
+## Fields
+
+| Field | Description |
+|---|---|
+| `id` | `<language>#<FineWeb2 id>` for human pages; `…\|level0` / `…\|level5` for AI versions |
+| `seed_id` | the human page the row belongs to |
+| `language`, `language_name` | FineWeb2 language code and name |
+| `build` | `9-language` or `24-language` |
+| `level` | null for the human page; 0 or 5 for AI versions |
+| `kind`, `prose_author`, `idea_author` | what the row is and who wrote its prose and ideas |
+| `generator` | `gemini-3.7-flash` for AI versions |
+| `format` | the page's format (WebOrganizer taxonomy, one of the eight long-form formats) |
+| `words`, `text` | the document |
+| `seed_crawl_dump`, `seed_crawl_date` | Common Crawl snapshot and crawl date of the human page |
+| `date_filtered_build` | whether the page's build was restricted to crawls before 2022 |
+
+## Caveats
+
+- A crawl date after 2022 does not show that a page was AI-written, but such pages can be. Their human label is
+  less certain.
+- URLs are not included. The FineWeb2 id identifies each page in the source dataset.
+
+## License
+
+The human pages come from FineWeb2 (ODC-By 1.0) and remain subject to the terms of their original sources and of
+Common Crawl. The AI versions were produced with Google Gemini models and are subject to Google's terms. Annotations
+are released under CC BY 4.0.
+
+## Citation
+
+```bibtex
+@article{idealens2026,
+  title   = {IdeaLens: Detecting AI Ideas in Long-form Writing},
+  author  = {Anonymous},
+  journal = {arXiv preprint arXiv:TBD},
+  year    = {2026},
+  url     = {https://arxiv.org/abs/TBD}
+}
+```

data/ideashift_x.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:020303e04d559d9ae2d4644544d3be51e85834ce8e3673056605b136dd4cdd5d
+size 211374287