Skip to content

Latest commit

 

History

History
15 lines (12 loc) · 1 KB

File metadata and controls

15 lines (12 loc) · 1 KB

Reproducible coverage benchmark

The scanner performs deterministic literal-character matching. Its benchmark goal is exact reporting and conservative cleaning, not AI authorship classification.

Category Expected behavior
Zero-width characters Report and remove supported code points
Directional controls Report and remove supported code points
Unicode tag characters Report and remove supported code points
Soft hyphens and byte-order marks Report and remove
Special spaces and invisible fillers Report and normalize conservatively
Mixed-script lookalikes Report for review and do not rewrite
Statistical provider watermarks Return no authorship conclusion

Run npm test to reproduce the included fixtures. The versioned public fixture set and rendered results are available at https://aitextwatermark.com/benchmarks/artifact-coverage. Browser answer-selection fixtures and limitations are published at https://aitextwatermark.com/extension/compatibility.