Triplet Evaluation (Ori-Imp) on Intent Semantic Toolkit (test)
71Teasy Scoreinstructor-large
Evaluation results
| Method | Links | ||
|---|---|---|---|
| instructor-large2024.03 | 71 | 0.036 | |
| paraphrase2024.03 | 68.9 | 0.039 | |
| instructor-base2024.03 | 68 | 0.02 | |
| IAE2024.03 | 67.3 | 0.032 | |
| Ours bestbackbone=instructor-large, configuration=-P4, N1,32024.03 | 20.4 | 0.776 | |
| Disable LLMbackbone=instructor-large, LLM-generated data=false2024.03 | 8 | 0.508 | |
| Baselinebackbone=instructor-large2024.03 | 3.3 | 0.706 |