Claim Verification on FEVEROUS
0.9567AccuracyInfoRE + CoT
Evaluation Results
| Method | Links | |
|---|---|---|
| InfoRE + CoTBackbone=GPT-4, Zero-shot=true2024.04 | 0.9567 | |
| InfoREBackbone=GPT-4, Zero-shot=true2024.04 | 0.9562 | |
| CoTBackbone=GPT-4, Zero-shot=true2024.04 | 0.9267 | |
| StandardBackbone=GPT-4, Zero-shot=true2024.04 | 0.9233 | |
| InfoRE + CoTBackbone=GPT-3.5, Zero-shot=true2024.04 | 0.9153 | |
| InfoREBackbone=GPT-3.5, Zero-shot=true2024.04 | 0.9131 | |
| CoTBackbone=GPT-3.5, Zero-shot=true2024.04 | 0.8867 | |
| StandardBackbone=GPT-3.5, Zero-shot=true2024.04 | 0.8767 | |
| InfoRE + CoTBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 0.6812 | |
| InfoREBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 0.6784 | |
| CoTBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 0.6453 | |
| StandardBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 0.6339 |