spb/wpai0002_uqo

1| system | n | chrf | edit_sim | rougeL | exact_match | number_recall | placeholders_exact |2|---|---|---|---|---|---|---|---|3| B0_identity | 531 | 74.097 [72.985, 75.536] | 0.730 [0.721, 0.742] | 0.735 [0.723, 0.751] | 0.000 [0.000, 0.000] | 1.000 [1.000, 1.000] | 1.000 [1.000, 1.000] |4| B1_generic_paraphrase | 531 | 70.564 [69.491, 71.968] | 0.597 [0.584, 0.611] | 0.640 [0.627, 0.655] | 0.000 [0.000, 0.000] | 1.000 [0.999, 1.000] | 1.000 [1.000, 1.000] |5| htr-m1-small | 531 | 73.015 [71.989, 74.328] | 0.717 [0.707, 0.730] | 0.723 [0.711, 0.737] | 0.000 [0.000, 0.000] | 0.993 [0.987, 0.997] | 0.994 [0.984, 1.000] |6| htr-m1-tiny-s3 | 531 | 73.246 [72.264, 74.465] | 0.723 [0.714, 0.734] | 0.726 [0.715, 0.739] | 0.000 [0.000, 0.000] | 0.991 [0.983, 0.997] | 0.992 [0.978, 1.000] |7| htr-m1-tiny | 531 | 73.125 [72.124, 74.394] | 0.721 [0.711, 0.732] | 0.725 [0.714, 0.739] | 0.000 [0.000, 0.000] | 0.993 [0.988, 0.996] | 0.991 [0.971, 1.000] |8| htr-m3tokenonly-tiny | 531 | 73.217 [72.243, 74.488] | 0.722 [0.713, 0.733] | 0.727 [0.716, 0.741] | 0.000 [0.000, 0.000] | 0.992 [0.987, 0.996] | 0.994 [0.984, 1.000] |910Stylometric W1 (mean over features, standardised; lower = closer to originals):11- rewrite vs original: 0.16312- B0_identity vs original: 0.163 · probe P(original) mean 0.4213- B1_generic_paraphrase vs original: 0.285 · probe P(original) mean 0.4214- htr-m1-small vs original: 0.102 · probe P(original) mean 0.5115- htr-m1-tiny-s3 vs original: 0.108 · probe P(original) mean 0.5016- htr-m1-tiny vs original: 0.110 · probe P(original) mean 0.5117- htr-m3tokenonly-tiny vs original: 0.111 · probe P(original) mean 0.5018