Fact-checking on FeLMWk
0.79F1 (True)PCC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PCCLLM=GPT-4o2026.01 | 0.79 | 0.68 | |
| PCCLLM=Gemini-2.5-Pro2026.01 | 0.77 | 0.74 | |
| FIRELLM=GPT-4o2026.01 | 0.77 | 0.63 | |
| PCCLLM=Gemini-2.5-Flash2026.01 | 0.76 | 0.72 | |
| PCCLLM=GPT-4o-mini2026.01 | 0.76 | 0.6 | |
| SAFELLM=GPT-4o2026.01 | 0.75 | 0.65 | |
| FIRELLM=Gemini-2.5-Pro2026.01 | 0.74 | 0.71 | |
| FIRELLM=Gemini-2.5-Flash2026.01 | 0.73 | 0.7 | |
| FIRELLM=GPT-4o-mini2026.01 | 0.71 | 0.53 | |
| SAFELLM=GPT-4o-mini2026.01 | 0.68 | 0.51 | |
| Factcheck-GPTLLM=GPT-4o2026.01 | 0.67 | 0.61 | |
| PCCLLM=Mistral-7B-Instruct2026.01 | 0.6 | 0.55 | |
| FactoolLLM=GPT-4o2026.01 | 0.6 | 0.64 | |
| FIRELLM=Mistral-7B-Instruct2026.01 | 0.56 | 0.52 | |
| Factcheck-GPTLLM=GPT-4o-mini2026.01 | 0.55 | 0.56 | |
| FactoolLLM=GPT-4o-mini2026.01 | 0.48 | 0.62 |