Scientific Reasoning on SciBench Chemistry
73.2AccuracySingle Agent
Evaluation results
| Method | Links | |
|---|---|---|
| Single AgentBase Model=GPT-4o, Aligned=false, AEA Model=None2026.05 | 73.2 | |
| Weak-to-Strong (AEA)Base Model=GPT-4o, Aligned=true, AEA Model=AEA-4B2026.05 | 73.2 | |
| Weak-to-Strong (AEA)Base Model=Claude 3.7 Sonnet, Aligned=true, AEA Model=AEA-4B2026.05 | 71.4 | |
| Multi AgentBase Model=GPT-4o, Aligned=false, AEA Model=None2026.05 | 70.7 | |
| Weak-to-Strong (AEA)Base Model=GPT-4o-mini, Aligned=true, AEA Model=AEA-4B2026.05 | 65.9 | |
| Self-Reflection (AEA)Base Model=GPT-4o, Aligned=true, AEA Model=GPT-4o2026.05 | 65.9 | |
| Self-Reflection (AEA)Base Model=Claude 3.7 Sonnet, Aligned=true, AEA Model=Claude 3.7 Sonnet2026.05 | 63.4 | |
| Single AgentBase Model=Claude 3.7 Sonnet, Aligned=false, AEA Model=None2026.05 | 61.5 | |
| Multi AgentBase Model=Claude 3.7 Sonnet, Aligned=false, AEA Model=None2026.05 | 61 | |
| Single AgentBase Model=GPT-4o-mini, Aligned=false, AEA Model=None2026.05 | 61 | |
| Multi AgentBase Model=GPT-4o-mini, Aligned=false, AEA Model=None2026.05 | 58.5 | |
| Weak-to-Strong (AEA)Base Model=Claude 3.5 Sonnet, Aligned=true, AEA Model=AEA-4B2026.05 | 56.1 | |
| Self-Reflection (AEA)Base Model=GPT-4o-mini, Aligned=true, AEA Model=GPT-4o-mini2026.05 | 55.9 | |
| Self-Reflection (AEA)Base Model=Claude 3.5 Sonnet, Aligned=true, AEA Model=Claude 3.5 Sonnet2026.05 | 54.9 | |
| Multi AgentBase Model=Claude 3.5 Sonnet, Aligned=false, AEA Model=None2026.05 | 51.2 | |
| Single AgentBase Model=Claude 3.5 Sonnet, Aligned=false, AEA Model=None2026.05 | 47.6 | |
| Weak-to-Strong (AEA)Base Model=Claude 3 Haiku, Aligned=true, AEA Model=AEA-4B2026.05 | 36.6 | |
| Multi AgentBase Model=Claude 3 Haiku, Aligned=false, AEA Model=None2026.05 | 33.6 | |
| Single AgentBase Model=Claude 3 Haiku, Aligned=false, AEA Model=None2026.05 | 31.7 | |
| Self-Reflection (AEA)Base Model=Claude 3 Haiku, Aligned=true, AEA Model=Claude 3 Haiku2026.05 | 31.7 |