Claim Verification on HOVER 4-hop
73.62AccuracyInfoRE + CoT
Evaluation Results
| Method | Links | |
|---|---|---|
| InfoRE + CoTBackbone=GPT-4, Zero-shot=true2024.04 | 73.62 | |
| InfoREBackbone=GPT-4, Zero-shot=true2024.04 | 73.08 | |
| CoTBackbone=GPT-4, Zero-shot=true2024.04 | 70.68 | |
| StandardBackbone=GPT-4, Zero-shot=true2024.04 | 70.06 | |
| InfoRE + CoTBackbone=GPT-3.5, Zero-shot=true2024.04 | 65.66 | |
| InfoREBackbone=GPT-3.5, Zero-shot=true2024.04 | 64.91 | |
| CoTBackbone=GPT-3.5, Zero-shot=true2024.04 | 62.69 | |
| StandardBackbone=GPT-3.5, Zero-shot=true2024.04 | 61.54 | |
| InfoRE + CoTBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 50.15 | |
| InfoREBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 50.04 | |
| CoTBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 48.01 | |
| StandardBackbone=LLAMA2-70B, Zero-shot=true2024.04 | 47.82 |