Instruction-following clustering on LMSY (C0)
66.9V-measureC1-Qwen-14B
Evaluation Results
| Method | Links | |
|---|---|---|
| C1-Qwen-14BModel Type=Our Model2026.03 | 66.9 | |
| C1-Qwen-7BModel Type=Our Model2026.03 | 65.63 | |
| o3Model Type=Reasoning Model2026.03 | 63.6 | |
| Gemini 2.5 ProModel Type=Reasoning Model2026.03 | 55.98 | |
| GPT-oss-120BModel Type=Reasoning Model2026.03 | 48.74 | |
| QwQ-32BModel Type=Reasoning Model2026.03 | 47.51 | |
| DeepSeek-R1Model Type=Reasoning Model2026.03 | 42.96 | |
| Distill-Llama-70BModel Type=Reasoning Model2026.03 | 37.8 | |
| GPT-4.1Model Type=General Model2026.03 | 37.27 | |
| GPT-4oModel Type=General Model2026.03 | 26.95 | |
| Distill-Qwen-32BModel Type=Reasoning Model2026.03 | 26.5 | |
| Llama-3.1-70B-InstructModel Type=General Model2026.03 | 24.57 | |
| Qwen2.5-72B-InstructModel Type=General Model2026.03 | 18.85 |