Question Answering on SQuAD v2 (Corr., Err.)
1,853CorrelationLLM-RAG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.142026.05 | 1,853 | 11.93 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 1,853 | 11.93 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.132026.05 | 1,850 | 11.75 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 1,843 | 12.4 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.142026.05 | 1,842 | 12.41 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 1,837 | 12.69 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.132026.05 | 1,836 | 12.1 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.142026.05 | 1,835 | 12.63 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.122026.05 | 1,831 | 10.95 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.122026.05 | 1,819 | 11.18 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.112026.05 | 1,803 | 10 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.112026.05 | 1,784 | 10.1 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 1,755 | 14.16 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 1,745 | 13.75 | |
| UCB-CascadedBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 1,744 | 13.54 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.142026.05 | 1,714 | 13.16 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.142026.05 | 1,699 | 12.71 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.132026.05 | 1,666 | 12.16 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.132026.05 | 1,654 | 11.64 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.122026.05 | 1,617 | 11.12 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.122026.05 | 1,591 | 10.66 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.112026.05 | 1,546 | 10.08 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.112026.05 | 1,484 | 9.75 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 211 | 13.24 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.142026.05 | 194 | 13.19 | |
| LLM-onlyBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 137 | 13.45 |