Problem Solving and Unsolvability Detection on Hitori
98Accuracy (Solvable)Gemini-3
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-3Model Scale=32025.12 | 98 | 100 | 99 | |
| Deepseek-V3.2-RModel Scale=V3.2-R2025.12 | 70 | 86 | 78 | |
| Qwen3-4B + UnsolvableRLModel Scale=4B, Training Protocol=UnsolvableRL2025.12 | 63.5 | 94.5 | 79 | |
| Qwen3-4B InstructModel Scale=4B, Training Protocol=Instruct2025.12 | 34.5 | 6.5 | 20.5 | |
| Qwen3-1.7B + UnsolvableRLModel Scale=1.7B, Training Protocol=UnsolvableRL2025.12 | 11 | 59 | 35 | |
| Qwen3-1.7B InstructModel Scale=1.7B, Training Protocol=Instruct2025.12 | 8 | 21 | 14.5 | |
| GPT-5.1-LowModel Scale=5.1-Low2025.12 | 6 | 12 | 9 |