Knowledge Base Question Answering on WebQSP -> GrailQA-Tech (test)
74.6F1 ScoreFuSIC-KBQA
Evaluation Results
| Method | Links | |
|---|---|---|
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=TIARA and Pangu2023.11 | 74.6 | |
| FUn-FuSIC(sc)prior knowledge of answerability=true2024.06 | 73.6 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=TIARA and Pangu2023.11 | 71.4 | |
| FUn-FuSIC(scUn)prior knowledge of answerability=false2024.06 | 71.2 | |
| FuSIC-KBQAprior knowledge of answerability=true2024.06 | 70.8 | |
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=TIARA2023.11 | 70.8 | |
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=Pangu2023.11 | 70.3 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=TIARA2023.11 | 68.4 | |
| KB-BinderEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 68 | |
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=TIARA and Pangu2023.11 | 65.1 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=Pangu2023.11 | 65.1 | |
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=TIARA2023.11 | 65 | |
| FuSIC-KBQA-Uprior knowledge of answerability=false2024.06 | 62.6 | |
| FuSIC-KBQAEvaluation Protocol=Few-shot, Category=Fusion, Retriever=Pangu2023.11 | 61.9 | |
| gf-LLMEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 61.2 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=TIARA and Pangu2023.11 | 56.6 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=TIARA2023.11 | 55.9 | |
| KB-BinderEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 55.7 | |
| gf-LLMEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 53.4 | |
| FuSIC-KBQAEvaluation Protocol=Zero-shot, Category=Fusion, Retriever=Pangu2023.11 | 52.5 | |
| PanguEvaluation Protocol=Few-shot, Category=Supervised2023.11 | 51 | |
| TIARAEvaluation Protocol=Few-shot, Category=Supervised2023.11 | 48.2 | |
| PanguEvaluation Protocol=Few-shot, Category=Supervised2023.11 | 45.3 | |
| gf-LLMEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 42.8 | |
| PanguEvaluation Protocol=Zero-shot, Category=Supervised2023.11 | 40.8 | |
| TIARAEvaluation Protocol=Few-shot, Category=Supervised2023.11 | 40.6 | |
| PanguEvaluation Protocol=Zero-shot, Category=Supervised2023.11 | 39.3 | |
| KB-BinderEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 39 | |
| BERT-RankingEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 35.9 | |
| BERT-RankingEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 34.4 | |
| KB-BinderEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 33.6 | |
| BERT-RankingEvaluation Protocol=Few-shot, Category=LLM ICL2023.11 | 33.1 | |
| gf-LLMEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 31.3 | |
| TIARAEvaluation Protocol=Zero-shot, Category=Supervised2023.11 | 30.6 | |
| TIARAEvaluation Protocol=Zero-shot, Category=Supervised2023.11 | 30 | |
| BERT-RankingEvaluation Protocol=Zero-shot, Category=LLM ICL2023.11 | 21.5 |