Knowledge Graph Question Answering on WebQSP 1-hop
76.9AccuracyCa2KG
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ca2KGBase Model=GPT-3.52026.01 | 76.9 | 0.196 | 0.184 | 0.837 | |
| Ca2KGBase Model=LLaMA-32026.01 | 73.8 | 0.211 | 0.203 | 0.812 | |
| RC-RAGBase Model=GPT-3.52026.01 | 70.4 | 0.302 | 0.295 | 0.727 | |
| RC-RAGBase Model=LLaMA-32026.01 | 64.2 | 0.269 | 0.258 | 0.759 | |
| Verb2S-Top4Base Model=LLaMA-32026.01 | 64.2 | 0.269 | 0.258 | 0.759 | |
| Verb1S-Top4Base Model=GPT-3.52026.01 | 61.8 | 0.309 | 0.313 | 0.724 | |
| KG-RAGBase Model=LLaMA-32026.01 | 57.9 | 0.33 | 0.301 | 0.71 | |
| IoEBase Model=GPT-3.52026.01 | 55.7 | — | — | — | |
| Verb2S-CoTBase Model=GPT-3.52026.01 | 54.8 | 0.42 | 0.401 | 0.593 | |
| Verb2S-Top4Base Model=GPT-3.52026.01 | 45.8 | 0.291 | 0.285 | 0.731 | |
| Self-CorrectBase Model=LLaMA-32026.01 | 45.5 | — | — | — | |
| Self-CorrectBase Model=GPT-3.52026.01 | 40.9 | — | — | — | |
| IoEBase Model=LLaMA-32026.01 | 37.2 | — | — | — | |
| Verb1S-Top4Base Model=LLaMA-32026.01 | 35.8 | 0.456 | 0.436 | 0.544 | |
| KG-RAGBase Model=GPT-3.52026.01 | 34 | 0.646 | 0.538 | 0.529 | |
| Verb2S-CoTBase Model=LLaMA-32026.01 | 32.4 | 0.63 | 0.577 | 0.433 |