Reasoning-intensive classification on StereoSet (test)
93Macro F1 ScoreClustered Instructions (PLD)
Evaluation Results
| Method | Links | |
|---|---|---|
| Clustered Instructions (PLD)Model=Gemini 3 Flash (Teacher)2026.02 | 93 | |
| Post Conflict resolution (PLD)Model=Gemini 2 Flash2026.02 | 93 | |
| Post Conflict resolution (PLD)Model=Gemini 3 Flash (Teacher)2026.02 | 93 | |
| Zero-shotModel=Gemini 3 Flash (Teacher)2026.02 | 92 | |
| Clustered Instructions (PLD)Model=Gemini 2 Flash2026.02 | 92 | |
| Few-shotModel=Gemini 3 Flash (Teacher)2026.02 | 90 | |
| Clustered Instructions (PLD)Model=Gemma2026.02 | 90 | |
| Post Conflict resolution (PLD)Model=Gemma2026.02 | 90 | |
| Few-shotModel=Gemini 2 Flash2026.02 | 83 | |
| Few-shotModel=Gemma2026.02 | 75 | |
| Zero-shotModel=Gemma2026.02 | 57 | |
| Zero-shotModel=Gemini 2 Flash2026.02 | 53 |