Question Answering on NQ (Corr. and Err.)
763Correlation ScoreLLM-RAG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.182026.05 | 763 | 14.78 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.192026.05 | 763 | 14.78 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.172026.05 | 760 | 14.66 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.192026.05 | 759 | 15.17 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.182026.05 | 757 | 15.23 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.172026.05 | 753 | 15.11 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.162026.05 | 752 | 14.22 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.192026.05 | 747 | 15.81 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.182026.05 | 743 | 14.92 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.162026.05 | 741 | 14.2 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.172026.05 | 736 | 14.08 | |
| LLM-RAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 734 | 13.31 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.162026.05 | 718 | 13.07 | |
| BalanceRAGBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 709 | 13.15 | |
| UCB-CascadedBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 698 | 12.14 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.192026.05 | 615 | 16.97 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.192026.05 | 615 | 16.62 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.182026.05 | 590 | 16.01 | |
| UCB-CascadedBackbone=Llama3.2-3B, Target risk level (alpha)=0.192026.05 | 590 | 15.82 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.182026.05 | 575 | 15.06 | |
| UCB-CascadedBackbone=Llama3.2-3B, Target risk level (alpha)=0.182026.05 | 563 | 15.2 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.172026.05 | 562 | 14.97 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.162026.05 | 532 | 13.9 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.172026.05 | 531 | 14.04 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.162026.05 | 492 | 13.12 | |
| LLM-RAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 489 | 12.83 | |
| BalanceRAGBackbone=Llama3.2-3B, Target risk level (alpha)=0.152026.05 | 447 | 12.2 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.192026.05 | 320 | 16.3 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.182026.05 | 283 | 14.93 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.172026.05 | 240 | 13.62 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.162026.05 | 191 | 11.84 | |
| LLM-onlyBackbone=Llama3.1-8B, Target risk level (alpha)=0.152026.05 | 158 | 10.55 | |
| LLM-onlyBackbone=Llama3.2-3B, Target risk level (alpha)=0.192026.05 | 134 | 15.72 | |
| LLM-onlyBackbone=Llama3.2-3B, Target risk level (alpha)=0.182026.05 | 125 | 15.52 | |
| LLM-onlyBackbone=Llama3.2-3B, Target risk level (alpha)=0.172026.05 | 119 | 15.68 |