Question Answering (Truthfulness) on TruthfulQA (Accuracy)
72.05AccuracyAgentRevive
Evaluation Results
| Method | Links | |
|---|---|---|
| AgentReviveVar. NS=True, Flex. State=True, Base model=Qwen2.5-72B-Instruct2026.05 | 72.05 | |
| AgentDropoutVar. NS=True, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 69.11 | |
| ARG-DesignerVar. NS=True, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 68.02 | |
| G-DesignerVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 67.55 | |
| AgentPruneVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 67.41 | |
| AgentVerseVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 66.64 | |
| AutogenVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 65.89 | |
| MASround=TVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 65.76 | |
| SC (CoT)Var. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 64.85 | |
| CoTVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 64.61 | |
| MASround=1Var. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 63.98 | |
| VanillaVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.05 | 62.98 |