Claim Verification on ExClaim
34Precision (P)CO-FACTCHECKER
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| CO-FACTCHECKERSystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle2026.04 | 34 | 27 | 28 | 21 | 87 | 36 | |
| Verifier - MCTSSystem Category=Autonomous, Search/Refinement Strategy=MCTS2026.04 | 31 | 25 | 25 | 20 | 86 | 30 | |
| Multi-Turn DialogueSystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle2026.04 | 31 | 26 | 25 | 20 | 86 | 23 | |
| CO-FACTCHECKER - w/o reward modelSystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle, Ablation Setting=w/o reward model2026.04 | 31 | 26 | 27 | 20 | 86 | 32 | |
| CO-FACTCHECKER - w/o guidance tokensSystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle, Ablation Setting=w/o guidance tokens2026.04 | 31 | 25 | 26 | 19 | 83 | 30 | |
| Verifier - best-of-NSystem Category=Autonomous, Search/Refinement Strategy=best-of-N2026.04 | 30 | 25 | 25 | 20 | 86 | 27 | |
| CO-FACTCHECKER - w/o remove / modifySystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle, Ablation Setting=w/o remove / modify2026.04 | 30 | 26 | 26 | 20 | 86 | 28 | |
| VerifierSystem Category=Autonomous2026.04 | 29 | 24 | 25 | 20 | 86 | 28 | |
| Verifier - self-refineSystem Category=Autonomous, Search/Refinement Strategy=self-refine2026.04 | 29 | 24 | 24 | 19 | 85 | 24 | |
| Choose OneSystem Category=Simulated Human-AI Collaboration, Oracle Simulation=Using an Oracle2026.04 | 29 | 25 | 25 | 19 | 86 | 28 | |
| SAFESystem Category=Autonomous2026.04 | 26 | 16 | 17 | 19 | 86 | 28 | |
| FIRESystem Category=Autonomous2026.04 | 24 | 16 | 16 | 18 | 86 | 28 | |
| Deep ResearchSystem Category=Autonomous2026.04 | 22 | 31 | 24 | 16 | 83 | — | |
| FactCheckGPTSystem Category=Autonomous2026.04 | 21 | 16 | 16 | 19 | 85 | 27 |