Question Answering on StrategyQA (ACC, JCD, F&C, C&A, ARS)
85.67Accuracy (ACC)GEMINI-3.1-PRO
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| GEMINI-3.1-PRO2026.05 | 85.67 | 87.75 | 90.13 | 82.04 | 87.17 | |
| GPT-5.12026.05 | 84.67 | 86.78 | 90.4 | 82.77 | 87.88 | |
| DEEPSEEK-R12026.05 | 84 | 86.13 | 89.92 | 82.73 | 87.77 | |
| Mixture-of-Agents2026.05 | 83.67 | 89.47 | 91.77 | 85.35 | 89.82 | |
| Self-Detection2026.05 | 82.33 | 89.02 | 90.72 | 84.55 | 89.83 | |
| LMVSLM2026.05 | 82 | 89.67 | 91.3 | 84.64 | 89.68 | |
| MAVEN2026.05 | 82 | 91.46 | 92.72 | 87.84 | 93.11 | |
| Multi-LLM Collab2026.05 | 81.67 | 90.12 | 91.48 | 86.08 | 89.83 | |
| DEEPSEEK-V3.22026.05 | 81.33 | 84.46 | 85.48 | 80.87 | 86.49 | |
| Self-Discover2026.05 | 81 | 88.75 | 90.78 | 83.88 | 89.52 | |
| ReConcile2026.05 | 78 | 80.92 | 86.23 | 76.44 | 86.37 |