Cultural norm reasoning on NormAd Africa 1.0 (test)
68.82AccuracyQwen3.5-4B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3.5-4BPrompt Language=Source-Language Stories2026.03 | 68.82 | |
| Tiny Aya FirePrompt Language=Source-Language Stories2026.03 | 66.28 | |
| Tiny Aya EarthPrompt Language=Source-Language Stories2026.03 | 66.19 | |
| Gemma3-4BPrompt Language=Source-Language Stories2026.03 | 65.46 | |
| Tiny Aya WaterPrompt Language=Source-Language Stories2026.03 | 65.37 | |
| Tiny Aya GlobalPrompt Language=Source-Language Stories2026.03 | 64.7 | |
| SmolLM3-3BPrompt Language=Source-Language Stories2026.03 | 61.52 | |
| Ministral-3-3BPrompt Language=Source-Language Stories2026.03 | 60.36 | |
| Qwen3-4BPrompt Language=Source-Language Stories2026.03 | 57.48 |