Language Adherence on MIF (target)
99.54RPRSFT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SFTEvaluation Setting=English as a Neutral Category2026.04 | 99.54 | 99.96 | |
| TLPOEvaluation Setting=English as a Neutral Category, N=162026.04 | 99.46 | 99.99 | |
| DPOEvaluation Setting=English as a Neutral Category2026.04 | 98.92 | 99.83 | |
| ORPOEvaluation Setting=English as a Neutral Category2026.04 | 98.03 | 99.91 | |
| BaselineEvaluation Setting=English as a Neutral Category2026.04 | 97.52 | 99.97 |