Question Answering on TruthfulQA (ACC, JCD, F&C, C&A, ARS)
93.67ACCMixture-of-Agents
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Mixture-of-Agents2026.05 | 93.67 | 86.42 | 90.32 | 86.64 | 89.97 | |
| GEMINI-3.1-PRO2026.05 | 92.33 | 89.43 | 91.37 | 88.11 | 93.13 | |
| DEEPSEEK-R12026.05 | 90 | 84.62 | 88.19 | 84.04 | 89.28 | |
| GPT-5.12026.05 | 90 | 87.53 | 90.76 | 87.54 | 92.3 | |
| MAVEN2026.05 | 90 | 90.77 | 90.9 | 89.37 | 96.38 | |
| Self-Discover2026.05 | 87.33 | 86.91 | 88.84 | 85.02 | 90.05 | |
| Multi-LLM Collab2026.05 | 87 | 87.32 | 89.07 | 86.18 | 91.57 | |
| Self-Detection2026.05 | 87 | 87.04 | 88.62 | 85.78 | 90.43 | |
| DEEPSEEK-V3.22026.05 | 85.67 | 82.82 | 87.52 | 83.48 | 85.97 | |
| LMVSLM2026.05 | 84.33 | 87.57 | 88.18 | 84.99 | 90.63 | |
| ReConcile2026.05 | 79.67 | 79.81 | 83.37 | 75.46 | 84.35 |