Evidence-Based Fact-Checking on AMBIFC
85.2AccuracySTRMAC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| STRMACBase LLM Model=Llama3.1-70B2025.11 | 85.2 | 338 | 82.4 | |
| MOABase LLM Model=Llama3.1-70B2025.11 | 83.2 | 6,004.4 | 45.6 | |
| MDAgentsBase LLM Model=Llama3.1-70B2025.11 | 82.2 | 35,165.7 | 2.4 | |
| IG-MASBase LLM Model=Llama3.1-70B2025.11 | 82.2 | 5,777 | 46.1 | |
| STRMACBase LLM Model=Llama3.1-8B2025.11 | 80.2 | 182.3 | 78.8 | |
| Random-ChainBase LLM Model=Llama3.1-70B2025.11 | 79.2 | 2,811.7 | 59.8 | |
| IG-MASBase LLM Model=Llama3.1-8B2025.11 | 79.2 | 2,371.4 | 62.5 | |
| MOABase LLM Model=Llama3.1-8B2025.11 | 73.3 | 6,427.2 | 38.5 | |
| MDAgentsBase LLM Model=Qwen2.5-32B2025.11 | 72.3 | 27,817.9 | 4.5 | |
| STRMACBase LLM Model=Qwen2.5-32B2025.11 | 72.3 | 486.2 | 68.9 | |
| MDAgentsBase LLM Model=Llama3.1-8B2025.11 | 71.3 | 44,514.3 | 0.8 | |
| Random-ChainBase LLM Model=Llama3.1-8B2025.11 | 65.4 | 2,996.1 | 48.5 | |
| LLM-DebateBase LLM Model=Qwen2.5-32B2025.11 | 65.4 | 6,686.8 | 33.5 | |
| IG-MASBase LLM Model=Qwen2.5-32B2025.11 | 58.4 | 2,303.3 | 46.4 | |
| LLM-DebateBase LLM Model=Llama3.1-70B2025.11 | 56.4 | 12,523.4 | 16.1 | |
| MOABase LLM Model=Qwen2.5-32B2025.11 | 51.5 | 6,169.9 | 27.8 | |
| Random-ChainBase LLM Model=Qwen2.5-32B2025.11 | 48.5 | 3,009.6 | 35.9 | |
| MACNETBase LLM Model=Llama3.1-70B2025.11 | 47.5 | 902.7 | 43.4 | |
| MACNETBase LLM Model=Qwen2.5-32B2025.11 | 47.5 | 1,124.3 | 42.4 | |
| MACNETBase LLM Model=Llama3.1-8B2025.11 | 46.5 | 960 | 42.2 | |
| LLM-DebateBase LLM Model=Llama3.1-8B2025.11 | 43.6 | 11,988.3 | 13.1 |