Reproducing benchmarks

by camillevanhoffelen - opened 2 days ago

I don't get the same benchmark results for Gretel or nemotron in my PII masking benchmark: https://huggingface.co/spaces/piimb/pii-masking-benchmark-leaderboard .
What does "18-locale-filtered" mean? Which ai4privacy dataset did you use? Which MAPA dataset did you use?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment