Card: link the idealens package and GitHub repo; full card

README.md

@@ -1,24 +1,73 @@
 ---
 license: apache-2.0
-tags: [ai-detection, text-classification]
+base_model: answerdotai/ModernBERT-large
+library_name: transformers
+language:
+- en
+tags:
+- ai-text-detection
+- idea-provenance
+datasets:
+- rishanthrajendhran/WildOutlines
 extra_gated_prompt: "Access is granted individually. Please say who you are and what you intend to use the weights for."
 ---
 
-# ideadet-modernbert-1m-roles
+# IdeaLens-ModernBERT-L-RolesOnly
 
-Idea-level AI-text detector. Answers **whose ideas a document contains**, not who typed the
-sentences: a document whose ideas are a person's is labelled human however much of the surface
-an AI produced.
+IdeaLens-ModernBERT-L-RolesOnly is an idea-level detector: it judges **whose ideas a document contains**, not who wrote its words, so a document whose ideas are a person's counts as human however much of its prose an AI wrote. It is one of the detectors released with [IdeaLens](https://huggingface.co/rishanthrajendhran/IdeaLens) and trained on the same data.
 
-- **Backbone:** modernbert
-- **Training corpus:** 1m
-- **Reads at inference:** the outline's role labels only, with the content removed
+| | |
+|---|---|
+| Model | ModernBERT-large with a sequence-classification head |
+| Reads | only the outline's sequence of role labels, one `[Role]` per line, with the content removed |
+| Training data | [WildOutlines](https://huggingface.co/datasets/rishanthrajendhran/WildOutlines), train split |
+| Output | P(human); a document is flagged as AI when P(human) is below a cut |
+| Default cut | 0.05082 (global, 1% false-positive rate) |
+| Hardware | any GPU; a CPU works for small jobs |
 
-Stored `y=1` means HUMAN. The score is `softmax(logits)[:, 0]` = P(human), and the detector
-fires when that falls below a calibrated threshold. Thresholds are quantiles of the score
-distribution over held-out human documents and are **not** included here; a cut from one
-model or input form is meaningless against another's scores.
+## Usage
 
-Weights are float32, exactly as used to produce the reported numbers.
+The [idealens](https://github.com/RishanthRajendhran/IdeaLens) package ([PyPI](https://pypi.org/project/idealens/)) runs the whole pipeline: it assigns each document one of the eight formats, extracts the outline with the prompt, role vocabulary and worked examples the detectors were trained with, and scores it with this model and the thresholds in this repo.
 
-Fuller documentation and evaluation results to follow.
+```bash
+pip install "idealens[hf]"
+idealens run docs.jsonl -o scores.jsonl --model IdeaLens-ModernBERT-L-RolesOnly
+```
+
+Input is JSONL with a `text` field per document. To score outlines you already have, use `idealens score outlines.jsonl -o scores.jsonl --model IdeaLens-ModernBERT-L-RolesOnly`. Score outlines as extracted; the paraphrasing step is only for training data.
+
+In Python:
+
+```python
+import idealens as il
+
+with il.Detector("IdeaLens-ModernBERT-L-RolesOnly") as det:
+    records = il.run(texts, det)   # classify, extract and score
+print(records[0]["p_human"], records[0]["verdict"]["ai"])
+```
+
+## Thresholds
+
+`thresholds.json` holds this model's cuts at 0.1%, 0.5%, 1%, 2% and 5% false-positive rates, fitted on the 80,000 human
+documents of WildOutlines' `calibration` split: one global cut per rate, plus per-format and per-topic cuts. The package applies them. A cut
+fitted for one model does not transfer to another model's scores. For documents unlike English web text, fit cuts on
+human documents from your own domain with `idealens calibrate`.
+
+## Related
+
+- [IdeaLens](https://huggingface.co/rishanthrajendhran/IdeaLens): the main idea-level detector, with full documentation
+- [ProseLens](https://huggingface.co/rishanthrajendhran/ProseLens): its prose-level counterpart
+- [idealens](https://github.com/RishanthRajendhran/IdeaLens): the Python package that runs these models
+- [WildOutlines](https://huggingface.co/datasets/rishanthrajendhran/WildOutlines): the training corpus
+
+## 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}
+}
+```