Fact-Checking on PolitiFact
72AccuracyToE
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| ToEBackbone=DeepSeek-V3.22026.06 | 72 | — | — | — | — | — | — | |
| TELLERBackbone=DeepSeek-V3.22026.06 | 68 | — | — | — | — | — | — | |
| TELLERBackbone=GPT-OSS-20B2026.06 | 64 | — | — | — | — | — | — | |
| ToEBackbone=GPT-OSS-20B2026.06 | 62 | — | — | — | — | — | — | |
| ZCoTBackbone=DeepSeek-V3.22026.06 | 60 | — | — | — | — | — | — | |
| STEELBackbone=DeepSeek-V3.22026.06 | 60 | — | — | — | — | — | — | |
| DefGenBackbone=GPT-OSS-20B2026.06 | 56 | — | — | — | — | — | — | |
| AFacTABackbone=GPT-OSS-20B2026.06 | 48 | — | — | — | — | — | — | |
| AdsentBackbone=GPT-OSS-20B2026.06 | 48 | — | — | — | — | — | — | |
| AdsentBackbone=DeepSeek-V3.22026.06 | 46 | — | — | — | — | — | — | |
| AFacTABackbone=DeepSeek-V3.22026.06 | 44 | — | — | — | — | — | — | |
| DirectBackbone=GPT-OSS-20B2026.06 | 40 | — | — | — | — | — | — | |
| DefGenBackbone=DeepSeek-V3.22026.06 | 38 | — | — | — | — | — | — | |
| STEELBackbone=GPT-OSS-20B2026.06 | 38 | — | — | — | — | — | — | |
| DirectBackbone=DeepSeek-V3.22026.06 | 32 | — | — | — | — | — | — | |
| ZCoTBackbone=GPT-OSS-20B2026.06 | 22 | — | — | — | — | — | — | |
| DeClarEMethod Category=Classical evidence based methods2026.02 | — | 65 | 68 | 67 | 65 | 61 | 66 | |
| EHIANMethod Category=Classical evidence based methods2026.02 | — | 67 | 68 | 65 | 65 | 62 | 62 | |
| FakeCheck-NoReSMethod Category=Ablation Study, Re-search=Disabled2026.02 | — | 80 | 79 | 82 | 79 | 80 | 78 | |
| FakeCheck-NoRetMethod Category=Ablation Study, Retrieval=Disabled2026.02 | — | 57 | 55 | 56 | 55 | 56 | 57 | |
| FakeCheckRAGBackbone=GPT-42026.02 | — | 85 | 83 | 86 | 83 | 84 | 83 | |
| FakeCheckRAG3.5Method Category=Ablation Study, Backbone=GPT-3.5-turbo2026.02 | — | 82 | 71 | 99 | 74 | 98 | 59 | |
| GETMethod Category=Classical evidence based methods2026.02 | — | 72 | 71 | 77 | 66 | 72 | 66 | |
| HANMethod Category=Classical evidence based methods2026.02 | — | 67 | 67 | 68 | 64 | 65 | 63 | |
| MACMethod Category=Classical evidence based methods2026.02 | — | 70 | 69 | 70 | 65 | 65 | 64 | |
| MUSERMethod Category=Classical evidence based methods2026.02 | — | 75 | 73 | 78 | 70 | 72 | 68 | |
| ProgramFCMethod Category=LLM-based methods2026.02 | — | 73 | 72 | 74 | 63 | 62 | 64 | |
| ReReadMethod Category=Classical evidence based methods2026.02 | — | 71 | 71 | 75 | 68 | 71 | 69 | |
| STEELMethod Category=LLM-based methods2026.02 | — | 78 | 74 | 78 | 72 | 74 | 72 | |
| Vicuna-7BMethod Category=LLM-based methods2026.02 | — | 52 | 53 | 52 | 51 | 52 | 51 | |
| WEBGLMMethod Category=LLM-based methods2026.02 | — | 60 | 61 | 63 | 61 | 66 | 62 |