release: publish SafeOCR-LabGold v0.1

README.md

@@ -0,0 +1,98 @@
+---
+pretty_name: SafeOCR-LabGold
+license: apache-2.0
+language:
+- en
+size_categories:
+- n<1K
+tags:
+- clinical-ocr
+- healthcare
+- synthetic
+- verification
+- fhir
+- benchmark
+---
+
+# SafeOCR-LabGold
+
+**SafeOCR-LabGold** is the synthetic laboratory-report benchmark used to develop and evaluate the frozen SafeOCR v0.1 verification policy.
+
+## Dataset summary
+
+The complete frozen split contains:
+
+| Split | Synthetic records | Rendered documents | Critical fields |
+|---|---:|---:|---:|
+| Calibration | 12 | 24 | 144 |
+| Final evaluation | 24 | 48 | 288 |
+| Total | 36 | 72 | 432 |
+
+Each synthetic record is rendered as a clean document and a deterministic corrupted counterpart. Three controlled templates are used: `classic`, `compact`, and `grid`. Every document contains six laboratory fields.
+
+The paper reports the **48-document / 288-critical-field final evaluation split**.
+
+## Intended use
+
+Use SafeOCR-LabGold for:
+
+- deterministic clinical-OCR verification research;
+- regression tests for evidence binding and selective-decision policies;
+- reproducible demonstrations of review/abstention behavior;
+- studying automation-versus-review trade-offs under the frozen v0.1 benchmark.
+
+Do **not** use this dataset as evidence of real-world clinical deployment performance. It is synthetic and does not represent the full distribution of real laboratory reports.
+
+## Frozen result boundary
+
+In the final evaluation split, the primary OCR baseline produced 0/288 observed errors. SafeOCR accepted 209/288 fields with 0/209 observed errors and routed 79 correct fields to review. Because the ungated baseline also had zero observed errors, this benchmark does **not** establish a safety benefit from gating.
+
+## Data generation and byte-level reproducibility
+
+The benchmark generator is deterministic for a fixed renderer stack:
+
+```bash
+python scripts/prepare_hf_labgold.py --output dist/SafeOCR-LabGold
+```
+
+The canonical Hugging Face v0.1 publication package is bound by:
+
+- `manifest_sha256 = 3a3fcdf1e3e045c0b6f8b334457d1e22dbd247a6a556e08235f808143abf5cec`;
+- Python 3.12.13;
+- Pillow 12.3.0;
+- FreeType 2.14.3;
+- JPEG feature version 8.0;
+- zlib 1.3.1.zlib-ng;
+- platform `win32:AMD64`.
+
+SafeOCR's corruption pipeline includes a JPEG round-trip. Pillow can therefore produce different corrupted PNG bytes when its underlying JPEG implementation differs across platforms even when the frozen seeds, case identities, layouts, and policy are unchanged. A package whose manifest hash differs from the canonical value above must not be presented as the byte-identical v0.1 publication artifact. For exact reproduction of the released benchmark, use the published Hugging Face artifacts or a matching renderer stack.
+
+This renderer dependency does not change the frozen evaluation results or claim boundaries; it is a byte-level reproducibility constraint on regenerated corrupted images.
+
+The generator records exact case identity, template, source seed, corruption seed, document hash, page hash, image filename, and truth-region metadata. It also writes Hugging Face `ImageFolder` metadata through `metadata.jsonl` with a `file_name` field so images and metadata can be loaded together.
+
+## Privacy
+
+All records are synthetic. Synthetic patient names and identifiers are generated by SafeOCR; no real patient data are used.
+
+## Repository and demo
+
+- Code and frozen evidence: https://github.com/AbdulazizShehri/SafeOCR
+- Interactive verification playground: https://huggingface.co/spaces/MedScaleAI/SafeOCR
+- Paper: arXiv identifier will be added at publication
+
+## Dataset contents
+
+The published repository contains:
+
+- `images/*.png`: deterministic synthetic laboratory report pages;
+- `metadata.jsonl`: Hugging Face `ImageFolder` metadata keyed by `file_name`;
+- `manifest.jsonl`: byte-identical audit manifest for the frozen release;
+- `dataset_info.json`: release counts and SHA-256 identities;
+- `README.md`: this dataset card.
+
+External datasets used in the paper are not redistributed here.
+
+## License
+
+The SafeOCR project and generated SafeOCR-LabGold artifacts are released under Apache-2.0 where applicable. Third-party datasets used in the paper retain their own terms and are **not** included in this Hugging Face dataset.

dataset_info.json

@@ -0,0 +1,10 @@
+{
+  "calibration_documents": 24,
+  "case_count": 72,
+  "critical_fields_per_document": 6,
+  "evaluation_documents": 48,
+  "manifest_sha256": "3a3fcdf1e3e045c0b6f8b334457d1e22dbd247a6a556e08235f808143abf5cec",
+  "metadata_sha256": "3a3fcdf1e3e045c0b6f8b334457d1e22dbd247a6a556e08235f808143abf5cec",
+  "readme_sha256": "c93570fc981740eb09617b0f93a5436f22a8694b411ad5587f44acc947516a23",
+  "schema_version": 1
+}