Binary decision on PLASMA
84.5AccuracyIDEA
Evaluation Results
| Method | Links | |
|---|---|---|
| IDEAModel=Qwen-3-32B2026.04 | 84.5 | |
| IDEAModel=Qwen-3-4B2026.04 | 83.9 | |
| CoTModel=GPT-5.22026.04 | 82.9 | |
| VanillaModel=Qwen-3-4B2026.04 | 82 | |
| IDEA w/o MCModel=Qwen-3-32B2026.04 | 82 | |
| VanillaModel=Qwen-3-32B2026.04 | 81.1 | |
| IDEA w/o InterModel=Qwen-3-8B2026.04 | 80.2 | |
| IDEAModel=Qwen-3-8B2026.04 | 80.2 | |
| BIRDModel=Qwen-3-32B2026.04 | 79.9 | |
| LogitModel=Qwen-3-32B2026.04 | 79.7 | |
| IDEA w/o MCModel=Qwen-3-8B2026.04 | 79.5 | |
| IDEA w/o InterModel=Qwen-3-4B2026.04 | 79.3 | |
| VanillaModel=Qwen-3-8B2026.04 | 78.1 | |
| IDEA w/o InterModel=Qwen-3-32B2026.04 | 76.8 | |
| IDEA w/o EMModel=Qwen-3-8B2026.04 | 76.6 | |
| LogitModel=Qwen-3-8B2026.04 | 76.4 | |
| IDEA w/o EMModel=Qwen-3-32B2026.04 | 76.3 | |
| CoTModel=Qwen-3-8B2026.04 | 75.9 | |
| IDEA w/o EMModel=Qwen-3-4B2026.04 | 75.7 | |
| LogitModel=Qwen-3-4B2026.04 | 75.6 | |
| CoTModel=DeepSeek R12026.04 | 75.4 | |
| BIRDModel=Qwen-3-4B2026.04 | 75.3 | |
| IDEA w/o MCModel=Qwen-3-4B2026.04 | 75.2 | |
| CoTModel=Qwen-3-4B2026.04 | 75.1 | |
| CoTModel=Qwen-3-32B2026.04 | 73.7 | |
| BIRDModel=Qwen-3-8B2026.04 | 73.3 | |
| RandomModel=N/A2026.04 | 50 |