Multi-hop Knowledge-based Question Answering on PathQuestion-Large (PQL) M 1.0
94.5AccuracyHypergraph Transformer
Evaluation results
| Method | Links | |
|---|---|---|
| Hypergraph TransformerSupervision=Weakly-supervised2022.04 | 94.5 | |
| SRNSupervision=Weakly-supervised2022.04 | 78.3 | |
| KV-MemNNSupervision=Fully-supervised2022.04 | 68.6 | |
| MINERVASupervision=Weakly-supervised2022.04 | 66.9 | |
| IRN-weakSupervision=Weakly-supervised2022.04 | 62.4 | |
| MemNNSupervision=Fully-supervised2022.04 | 55.8 |