Online Out-of-Distribution Detection on Num. Theory (Far-shift OOD)
92.08AccuracyTV Score
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TV ScoreModel=Llama2-7B, Evaluation Scenario=Online Detection2024.05 | 92.08 | 2.34 | |
| Input EmbeddingModel=Llama2-7B, Evaluation Scenario=Online Detection2024.05 | 85.8 | 3.31 | |
| Output EmbeddingModel=Llama2-7B, Evaluation Scenario=Online Detection2024.05 | 54.38 | 11.45 |