Upload README.md with huggingface_hub

README.md

@@ -1,60 +1,73 @@
 ---
 license: mit
-task_categories:
-- text-classification
-language:
-- en
-tags:
-- AI-Generated
-- Human-Generated
-- Academic
-pretty_name: AI-Generated Academic Text Detection
-size_categories:
-- 100K<n<1M
----
-language: en
-license: cc-by-4.0
 task_categories:
   - text-classification
+language:
+  - en
 tags:
   - ai-generated-text-detection
+  - machine-generated-text
   - academic-text
-  - distribution-shift
+size_categories:
+  - 100K<n<1M
+configs:
+  - config_name: default
+    data_files: academic_text.parquet
 ---
 
-# Academic Text Dataset
+# Academic Text: human vs AI-generated academic paragraphs
+
+669,008 academic paragraphs for AI-generated text detection in the STEM /
+academic domain.
+
+## Composition
+
+| source | label | rows |
+|---|---|---|
+| `arxiv` (papers published before 2022) | 0 (human) | 469,008 |
+| `gpt-3.5-turbo` | 1 (AI) | 100,000 |
+| `gemini-2.0-flash` | 1 (AI) | 100,000 |
+
+Human paragraphs are drawn from arXiv papers published before 2022, which
+predates widespread LLM authorship and so guarantees human origin. AI
+paragraphs were generated through the official APIs on STEM topics matched to
+the distribution of the human paragraphs.
+
+## Columns
 
-**A large-scale academic-domain benchmark for AI-generated text detection.**
+- `text` (string) -- the paragraph
+- `label` (int) -- 0 human, 1 AI-generated
+- `source` (string) -- `arxiv`, `gpt-3.5-turbo`, or `gemini-2.0-flash`
 
-This dataset was specifically constructed to evaluate robustness of AI-text detectors under realistic distribution shift in the academic (STEM) domain.
+## Usage
 
-## Dataset Description
+```python
+from datasets import load_dataset
+ds = load_dataset("mohamedmady/Academic-Text-arxiv-gpt-gemini")["train"]
+print(ds[0])
+```
 
-The dataset contains:
-- **469,008** human-written arXiv paragraphs (collected from papers published **before 2022**, guaranteeing purely human-generated text)
-- **100,000** AI-generated abstracts produced by **GPT-3.5 Turbo**
-- **100,000** AI-generated abstracts produced by **Gemini 2.0 Flash**
+## Notes
 
-**Total: 669,008 samples**
+The class balance is roughly 70 % human / 30 % AI. Report balanced accuracy or
+class-wise recall rather than raw accuracy.
 
-The AI texts were generated using the official APIs on STEM topics carefully matched to the distribution of the human arXiv abstracts, ensuring close semantic and topical alignment.
+The original blank-line-separated `.txt` files are retained for provenance, but
+the parquet is the canonical form: the text files carry no labels.
 
-## Usage in Our Paper
+## Licence
 
-We evaluate our best model (`DeBERTa-v3-base+FeatAttn`) under a **fixed-threshold protocol** (threshold calibrated once on HC3 PLUS validation and kept fixed).  
-On this dataset the model achieves:
-- Overall balanced accuracy: **81.09%**
-- GPT-3.5 Turbo detection: **100.0%**
-- Gemini 2.0 Flash detection: **93.25%**
-- Human (arXiv) recall: **75.57%**
+The generated paragraphs are released under MIT. The human paragraphs derive
+from arXiv, whose papers carry their own per-paper licences; check the terms of
+the source papers before redistributing that portion.
 
-## Recommended Citation
+## Citation
 
 ```bibtex
-@misc{mady2026academictext,
-  author       = {Mohamed Mady and Johannes Reschke and Björn Schuller},
-  title        = {Academic Text Dataset: A large-scale academic-domain benchmark for AI-generated text detection},
-  year         = {2026},
-  howpublished = {\url{https://huggingface.co/datasets/YOUR_USERNAME/academic-text-detection}},
+@misc{academictext2026,
+  title  = {Academic Text: A large-scale academic-domain benchmark for AI-generated text detection},
+  author = {Mady, Mohamed and Reschke, Johannes and Schuller, Bj\"orn},
+  year   = {2026},
+  howpublished = {Hugging Face dataset}
 }
----
+```