Veracity Assessment on BingCheck
0.88True F1MERMAID
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MERMAIDLLM=OSS-120B2026.01 | 0.88 | 0.68 | 0.78 | |
| MERMAIDLLM=GPT-4o2026.01 | 0.88 | 0.65 | 0.77 | |
| MERMAIDLLM=GPT-5-mini2026.01 | 0.88 | 0.7 | 0.79 | |
| SAFELLM=GPT-4o2026.01 | 0.87 | 0.65 | 0.76 | |
| FIRELLM=GPT-4o2026.01 | 0.87 | 0.63 | 0.76 | |
| MERMAIDLLM=Qwen-2.5-70B2026.01 | 0.87 | 0.61 | 0.74 | |
| MERMAIDLLM=Qwen-2.5-7B2026.01 | 0.85 | 0.61 | 0.73 | |
| MERMAIDLLM=OSS-20B2026.01 | 0.84 | 0.64 | 0.74 | |
| FactCheck-GPTLLM=GPT-4o2026.01 | 0.77 | 0.59 | 0.64 | |
| FacToolLLM=GPT-4o2026.01 | 0.68 | 0.56 | 0.62 | |
| MERMAIDLLM=LLaMA-3.1-70B-Inst2026.01 | 0.65 | 0.56 | 0.6 | |
| MERMAIDLLM=LLaMA-3.1-8B-Inst2026.01 | 0.6 | 0.46 | 0.53 |