Knowledge Base Question Answering 벤치마크
Knowledge Base Question Answering on ComplexWebQuestions
Accuracy
- 2021-01-11 — NSM+h: Accuracy 53.9
- 2021-04-18 — CBR-KBQA: Accuracy 70.4
- 2023-10-13 — ChatKBQA: Accuracy 76.8
| Rank | Model | Accuracy | F1 | Hits@1 | EM | Paper | Code | Year |
|---|---|---|---|---|---|---|---|---|
| 1 | ChatKBQA | 76.8 | 81.3 | 86.0 | – | ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Models | lhrlab/chatkbqa | 2023 |
| 2 | CBR-KBQA | 70.4 | – | – | – | Case-based Reasoning for Natural Language Queries over Knowledge Bases | 2021 | |
| 3 | NSM+h | 53.9 | – | – | – | Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals | RichardHGL/WSDM2021_NSM | 2021 |
| 4 | PullNet | 45.9 | – | – | – | Case-based Reasoning for Natural Language Queries over Knowledge Bases | 2021 | |
| 5 | QGG | 44.1 | – | – | – | Case-based Reasoning for Natural Language Queries over Knowledge Bases | 2021 | |
| 6 | PoG-GPT4 (Tan et al., 2024) | – | – | – | 81.4 | Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning | SteveTANTAN/PoG | 2024 |