Machine Translation Error Detection on WMT 25 (ESA)
17PrecisionAIP.sec
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AIP.secmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 17 | 14.75 | 15.17 | |
| AIP.primatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 16.06 | 13.02 | 13.52 | |
| AutoLQA41.secmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 11.68 | 7.27 | 8.3 | |
| AutoLQAESA.secmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 11.45 | 13.5 | 10.77 | |
| AutoLQA.primatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 10.61 | 14.59 | 10.95 | |
| XCOMET-XXL.basmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 6.8 | 28.45 | 10.55 | |
| XCOMET-XL.basmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 6.22 | 25.9 | 9.6 | |
| GemSpanEval-QE.secmatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 5.98 | 33.99 | 9.95 | |
| GemSpanEval.primatching_protocol=MPP, averaging_mode=micro-averaging2026.03 | 5.91 | 33.51 | 9.81 |