Kernekoncepter
Proposing a novel TKGC model, PPT, that leverages pre-trained language models with prompts to effectively incorporate temporal information from knowledge graphs.
Resumé
新しいTKGCモデル、PPTを提案。プリトレーニングされた言語モデルとプロンプトを活用して、知識グラフから時間情報を効果的に取り込む。
Statistik
ICEWS05-15: 368868 training instances, 46302 validation instances, 46159 test instances.
ICEWS14: 74845 training instances, 8514 validation instances, 7371 test instances.
ICEWS18: 373018 training instances, 45995 validation instances, 49545 test instances.
Citater
"Most existing methods focus on learning representations based on graph neural networks while inaccurately extracting information from timestamps and insufficiently utilizing the implied information in relations."
"Our model can effectively incorporate information from temporal knowledge graphs into the language models."