Knowledge Graph Question Answering on MetaQA 1-hop
87.6AccuracyCa2KG
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ca2KGBase Model=GPT-3.52026.01 | 87.6 | 0.067 | 0.078 | 0.947 | |
| RC-RAGBase Model=GPT-3.52026.01 | 87.4 | 0.112 | 0.094 | 0.934 | |
| Ca2KGBase Model=LLaMA-32026.01 | 86.1 | 0.085 | 0.093 | 0.944 | |
| IoEBase Model=GPT-3.52026.01 | 84.3 | — | — | — | |
| Verb2S-Top4Base Model=LLaMA-32026.01 | 83.4 | 0.147 | 0.118 | 0.921 | |
| Verb2S-Top4Base Model=GPT-3.52026.01 | 82.9 | 0.187 | 0.13 | 0.931 | |
| Verb1S-Top4Base Model=GPT-3.52026.01 | 81.3 | 0.16 | 0.152 | 0.897 | |
| RC-RAGBase Model=LLaMA-32026.01 | 81.2 | 0.124 | 0.118 | 0.912 | |
| IoEBase Model=LLaMA-32026.01 | 80.8 | — | — | — | |
| KG-RAGBase Model=LLaMA-32026.01 | 72.4 | 0.264 | 0.22 | 0.887 | |
| Verb2S-CoTBase Model=LLaMA-32026.01 | 66.8 | 0.373 | 0.282 | 0.844 | |
| Verb1S-Top4Base Model=LLaMA-32026.01 | 64.6 | 0.108 | 0.126 | 0.933 | |
| Self-CorrectBase Model=LLaMA-32026.01 | 60.5 | — | — | — | |
| Verb2S-CoTBase Model=GPT-3.52026.01 | 56.2 | 0.437 | 0.437 | 0.618 | |
| KG-RAGBase Model=GPT-3.52026.01 | 55.4 | 0.433 | 0.416 | 0.706 | |
| Self-CorrectBase Model=GPT-3.52026.01 | 53.9 | — | — | — |