Cultural norm reasoning on NormAd Americas 1.0 (test)
77.72AccuracyMinistral-3-3B
Evaluation Results
| Method | Links | |
|---|---|---|
| Ministral-3-3BPrompt Language=Source-Language Stories2026.03 | 77.72 | |
| SmolLM3-3BPrompt Language=Source-Language Stories2026.03 | 73.65 | |
| Qwen3.5-4BPrompt Language=Source-Language Stories2026.03 | 73.37 | |
| Tiny Aya FirePrompt Language=Source-Language Stories2026.03 | 73.02 | |
| Tiny Aya EarthPrompt Language=Source-Language Stories2026.03 | 70.71 | |
| Gemma3-4BPrompt Language=Source-Language Stories2026.03 | 70.25 | |
| Tiny Aya WaterPrompt Language=Source-Language Stories2026.03 | 69.4 | |
| Tiny Aya GlobalPrompt Language=Source-Language Stories2026.03 | 68.64 | |
| Qwen3-4BPrompt Language=Source-Language Stories2026.03 | 66.31 |