arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1806.00692cs.CL
收藏

自然语言推理的压力测试评估

Stress Test Evaluation for Natural Language Inference

Carnegie Mellon University(卡内基梅隆大学)

Aakanksha Naik, Abhilasha Ravichander, Norman Sadeh, Carolyn Rose, Graham Neubig

摘要:

自然语言推理(NLI)是确定自然语言假设是否可以以合理的方式从给定前提中推断出来的任务。NLI被提出作为自然语言理解的基准任务。现有的模型在NLI的标准数据集上表现良好,在不同类型的文本中取得了令人印象深刻的结果。然而,这些模型在多大程度上理解句子的语义内容尚不清楚。在这项工作中,我们提出了一种评估方法,包括自动构建的“压力测试”,使我们能够检查系统是否有能力作出真正的推理决定。我们对这些压力测试中的六个语音编码器模型的评估揭示了这些模型在具有挑战性的语言现象方面的优势和劣势,并为这一领域未来的工作提出了重要的方向。

英文摘要:

Natural language inference (NLI) is the task of determining if a natural language hypothesis can be inferred from a given premise in a justifiable manner. NLI was proposed as a benchmark task for natural language understanding. Existing models perform well at standard datasets for NLI, achieving impressive results across different genres of text. However, the extent to which these models understand the semantic content of sentences is unclear. In this work, we propose an evaluation methodology consisting of automatically constructed "stress tests" that allow us to examine whether systems have the ability to make real inferential decisions. Our evaluation of six sentence-encoder models on these stress tests reveals strengths and weaknesses of these models with respect to challenging linguistic phenomena, and suggests important directions for future work in this area.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑