zjunlp / KnowPrompt

[WWW 2022] KnowPrompt: Knowledge-aware Prompt-tuning with Synergistic Optimization for Relation Extraction
MIT License
194 stars 34 forks source link

遇到问题求助 #1

Closed zhangxq9944 closed 2 years ago

zhangxq9944 commented 2 years ago

您好,我想问一下您几个问题: 1: virtual template words 指的是构造的模板吗?也就是图一中的 Hamilton [MASK] British [SEP] 2: answer words 指的是 [MASK] 预测的的关系词吗? 图一中的 country,city,residence 3: 公式5中的 h,t,r 分别指的是什么? 4:文中多次提到 estimate the probability distributions ,这个值是怎么算的?

非常感谢

njcx-ai commented 2 years ago

您好:

  1. virtual template words 指的[sub]Hamilton [sub]其中的[sub],以避免歧义后更名叫virtual type words
  2. 是的
  3. 公式5中的h, t, r为s, o, r (这里描述笔误未与前文对齐)
  4. 根据pre-defined的类别算的,例如:per:birth_of_place. per:birth_of_data, 对于[sub],person: 1; 对于[sub],place: 0.5, data: 0.5