Binary Classification on GSS United States (test)
0.5636AccuracyOG-MAR
Evaluation Results
| Method | Links | |
|---|---|---|
| OG-MARBackbone=EXAONE 3.52026.01 | 0.5636 | |
| ValuesRAGBackbone=GPT-4o mini2026.01 | 0.5589 | |
| ValuesRAGBackbone=EXAONE 3.52026.01 | 0.552 | |
| DebateBackbone=GPT-4o mini2026.01 | 0.5509 | |
| OG-MARBackbone=Gemini 2.5 Flash Lite2026.01 | 0.5489 | |
| OG-MARBackbone=GPT-4o mini2026.01 | 0.548 | |
| DebateBackbone=EXAONE 3.52026.01 | 0.5407 | |
| ValuesRAGBackbone=Gemini 2.5 Flash Lite2026.01 | 0.5376 | |
| RoleBackbone=EXAONE 3.52026.01 | 0.5326 | |
| OG-MARBackbone=QWEN 2.52026.01 | 0.5325 | |
| Zero-shotBackbone=EXAONE 3.52026.01 | 0.5311 | |
| Self-consistencyBackbone=EXAONE 3.52026.01 | 0.5266 | |
| ValuesRAGBackbone=QWEN 2.52026.01 | 0.5215 | |
| RoleBackbone=GPT-4o mini2026.01 | 0.5184 | |
| DebateBackbone=QWEN 2.52026.01 | 0.5174 | |
| Zero-shotBackbone=GPT-4o mini2026.01 | 0.5164 | |
| DebateBackbone=Gemini 2.5 Flash Lite2026.01 | 0.5138 | |
| Zero-shotBackbone=QWEN 2.52026.01 | 0.5069 | |
| RoleBackbone=QWEN 2.52026.01 | 0.5037 | |
| RoleBackbone=Gemini 2.5 Flash Lite2026.01 | 0.4992 | |
| Self-consistencyBackbone=QWEN 2.52026.01 | 0.4975 | |
| Zero-shotBackbone=Gemini 2.5 Flash Lite2026.01 | 0.4957 | |
| Self-consistencyBackbone=GPT-4o mini2026.01 | 0.492 | |
| Self-consistencyBackbone=Gemini 2.5 Flash Lite2026.01 | 0.4728 |