Binary Classification on Intent Semantic Toolkit (test)
94.1Ori ScoreOurs best
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Ours bestBackbone=instructor-base, Prompt Variant=-P4, N1,32024.03 | 94.1 | 81 | 78.9 | |
| Ours bestbackbone=instructor-large, configuration=-P4, N1,32024.03 | 94 | 73.6 | 83.1 | |
| Baselinebackbone=instructor-large2024.03 | 89.6 | 73.9 | 79.4 | |
| Disable LLMbackbone=instructor-large, LLM-generated data=false2024.03 | 89.2 | 53 | 84.8 | |
| BaselineBackbone=instructor-base2024.03 | 89.1 | 67 | 78.4 | |
| Disable LLMBackbone=instructor-base2024.03 | 83.6 | 43.1 | 83.6 |