Instruction-following clustering on ECHR (C0)
85.75V-measureo3
Evaluation Results
| Method | Links | |
|---|---|---|
| o3Model Type=Reasoning Model2026.03 | 85.75 | |
| Gemini 2.5 ProModel Type=Reasoning Model2026.03 | 85.05 | |
| C1-Qwen-14BModel Type=Our Model2026.03 | 84.05 | |
| C1-Qwen-7BModel Type=Our Model2026.03 | 81.81 | |
| QwQ-32BModel Type=Reasoning Model2026.03 | 75.98 | |
| GPT-oss-120BModel Type=Reasoning Model2026.03 | 70.54 | |
| GPT-4.1Model Type=General Model2026.03 | 64.93 | |
| Distill-Llama-70BModel Type=Reasoning Model2026.03 | 63.45 | |
| GPT-4oModel Type=General Model2026.03 | 58.86 | |
| Distill-Qwen-32BModel Type=Reasoning Model2026.03 | 54.43 | |
| DeepSeek-R1Model Type=Reasoning Model2026.03 | 53.78 | |
| Qwen2.5-72B-InstructModel Type=General Model2026.03 | 45.57 | |
| Llama-3.1-70B-InstructModel Type=General Model2026.03 | 42.69 |