Uncertainty Estimation on MoreHopQA Camel
65.29AUROCMATU
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MATUBase LLM=Qwen2.5-7B2026.04 | 65.29 | 72.26 | |
| MATUBase LLM=Llama3.1-8B2026.04 | 63.2 | 65.61 | |
| P(true)Base LLM=Qwen2.5-7B2026.04 | 57.66 | 65.12 | |
| SAUP-MultipleBase LLM=Llama3.1-8B2026.04 | 56.68 | 57.98 | |
| Eigv(Agre)-finalBase LLM=Qwen2.5-7B2026.04 | 56.31 | 65.29 | |
| Eigv(Agre)-finalBase LLM=Llama3.1-8B2026.04 | 55.72 | 56.44 | |
| MATUBase LLM=GPT-4o2026.04 | 55.55 | 34.74 | |
| P(true)Base LLM=GPT-4o2026.04 | 54.8 | 34.05 | |
| Eigv(Agre)-WholeBase LLM=Qwen2.5-7B2026.04 | 54.2 | 63.42 | |
| Eigv(Agre)-WholeBase LLM=Llama3.1-8B2026.04 | 53.98 | 55.85 | |
| SAUP-MultipleBase LLM=Qwen2.5-7B2026.04 | 53.86 | 63.45 | |
| P(true)Base LLM=Llama3.1-8B2026.04 | 53.13 | 54.6 | |
| Eigv(Agre)-finalBase LLM=GPT-4o2026.04 | 53.07 | 33.74 | |
| Eigv(Agre)-WholeBase LLM=GPT-4o2026.04 | 52.59 | 33.19 | |
| SAUP-SingleBase LLM=Qwen2.5-7B2026.04 | 51.03 | 62.11 | |
| SAUP-SingleBase LLM=Llama3.1-8B2026.04 | 50.83 | 55.76 |