自然语言推理是NLP高级别的任务之一,不过自然语言推理包含的内容比较多,机器阅读,问答系统和对话等本质上都属于自然语言推理。最近在看AllenNLP包的时候,里面有个模块:文本蕴含任务(text entailment),它的任务形式是:

给定一个前提文本(premise),根据这个前提去推断假说文本(hypothesis)与premise的关系,一般分为蕴含关系(entailment)和矛盾关系(contradiction),蕴含关系(entailment)表示从premise中可以推断出hypothesis;矛盾关系(contradiction)即hypothesis与premise矛盾。文本蕴含的结果就是这几个概率值。

Textual Entailment
Textual Entailment (TE) models take a pair of sentences and predict whether the facts in the first necessarily imply the facts in the second one. The AllenNLP TE model is a re-implementation of the decomposable attention model (Parikh et al, 2017), a widely used TE baseline that was state-of-the-art onthe SNLI dataset in late 2016. The AllenNLP TE model achieves an accuracy of 86.4% on the SNLI 1.0 test dataset, a 2% improvement on most publicly available implementations and a similar score as the original paper. Rather than pre-trained Glove vectors, this model uses ELMo embeddings, which are completely character based and account for the 2% improvement.

AllenNLP集成了EMNLP2016中谷歌作者们撰写的一篇文章:A Decomposable Attention Model for Natural Language Inference

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

论文实践

(1)测试例子一:

前提:Two women are wandering along the shore drinking iced tea.

假设:Two women are sitting on a blanket near some rocks talking about politics.

其测试结果如下:

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

可视化呈现结果如下:

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

测试例子二:

前提:If you help the needy, God will reward you.

假设:Giving money to the poor has good consequences.

测试结果如下:

Textual Entailment(自然语言推理-文本蕴含) - AllenNLP-LMLPHP

05-14 03:56