Average Evaluation across 7 Tasks on General (test)
76.5BERTScoreTyphoon
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Typhoon2024.11 | 76.5 | 2.61 | 17.53 | 12.9 | 18.8 | 10 | 18.1 | 18 | 7.26 | |
| Seed-Free Synthetic Data Generation FrameworkFluency (F)=+, Culture (C)=+, Diversity (D)=+2024.11 | 73.2 | 0.08 | 9.56 | 6.7 | 6.6 | 2.7 | 6.4 | 6.5 | 4.58 | |
| WangchanX2024.11 | 72.2 | 2.08 | 17.24 | 11.1 | 14 | 7.1 | 13.3 | 13.4 | 7.43 | |
| OpenThai2024.11 | 67.5 | 0.88 | 14.84 | 8.4 | 10.4 | 4.4 | 9.9 | 9.9 | 3.31 | |
| Seed-Free Synthetic Data Generation FrameworkFluency (F)=+, Culture (C)=-, Diversity (D)=-2024.11 | 59.7 | 0.01 | 5.04 | 3.5 | 3 | 1.1 | 3 | 3 | 2.29 | |
| Seed-Free Synthetic Data Generation FrameworkFluency (F)=-, Culture (C)=+, Diversity (D)=-2024.11 | 52.6 | 0 | 2.82 | 1.7 | 1.5 | 0.2 | 1.5 | 1.5 | 0.83 | |
| Seed-Free Synthetic Data Generation FrameworkFluency (F)=-, Culture (C)=-, Diversity (D)=-2024.11 | 50.9 | 0.04 | 4.91 | 2.5 | 1.5 | 0.4 | 1.5 | 1.5 | 0.86 | |
| Seed-Free Synthetic Data Generation FrameworkFluency (F)=-, Culture (C)=-, Diversity (D)=+2024.11 | 49.5 | 0 | 2.56 | 1.6 | 1.6 | 0.3 | 1.6 | 1.6 | 0.73 |