Quality Estimation on WMT 24
0.606Pearson CorrelationSigmoid Head
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Sigmoid HeadModel=Olmo2026.01 | 0.606 | — | — | — | — | |
| Monte Carlo Seq. Entropy SigmoidModel=Olmo2026.01 | 0.552 | — | — | — | — | |
| Sigmoid HeadModel=Tower2026.01 | 0.468 | — | — | — | — | |
| Monte Carlo Seq. Entropy SigmoidModel=Tower2026.01 | 0.441 | — | — | — | — | |
| COMET KiwiEvaluation Mode=Eval Others2026.01 | 0.377 | 0.349 | 0.429 | 0.385 | 0.345 | |
| LLM Self JudgeModel=Tower2026.01 | 0.293 | — | — | — | — | |
| TowerQE Head=Sigmoid Head, Evaluation Mode=Eval Others2026.01 | 0.217 | 0.272 | 0.23 | 0.161 | 0.205 | |
| TowerQE Head=Boosted Prob, Evaluation Mode=Eval Others2026.01 | 0.17 | 0.215 | 0.17 | 0.114 | 0.18 | |
| Monte Carlo Seq. Entropy SoftmaxModel=Olmo2026.01 | 0.106 | — | — | — | — | |
| Monte Carlo Seq. Entropy SoftmaxModel=Tower2026.01 | 0.093 | — | — | — | — | |
| LLM Self JudgeModel=Olmo2026.01 | 0.093 | — | — | — | — | |
| TowerQE Head=Softmax Head, Evaluation Mode=Eval Others2026.01 | 0.058 | 0.094 | 0.056 | 0.011 | 0.072 |