HyperAI초신경

Kg To Text Generation On Webquestions

평가 지표

BLEU
METEOR
ROUGE

평가 결과

이 벤치마크에서 각 모델의 성능 결과

비교 표
모델 이름BLEUMETEORROUGE
jointgt-graph-text-joint-representation28.7830.5555.12
jointgt-graph-text-joint-representation29.6131.4855.42
toward-subgraph-guided-knowledge-graph29.4530.9655.45
jointgt-graph-text-joint-representation30.0232.0555.6
jointgt-graph-text-joint-representation28.9531.2954.47