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 1808.09132cs.CL
收藏

将自然语言命令映射到Web元素

Mapping Natural Language Commands to Web Elements

Stanford University(斯坦福大学)

Panupong Pasupat, Tian-Shun Jiang, Evan Zheran Liu, Kelvin Guu, Percy Liang

摘要:

网络提供了一个丰富的,开放的领域环境,具有文本,结构和空间属性。我们提出了一个新的任务,在这种环境中接地语言:给定一个自然语言命令(例如,"点击第二篇文章"),在网页上选择正确的元素(例如,超链接或文本框)。我们收集了超过50,000个命令的数据集,这些命令捕获了各种现象,如功能引用(例如“查找谁创建了这个网站”),关系推理(例如“约翰的文章”)和视觉推理(例如“最热门的文章”)。我们还实现并分析了三个基线模型,这些模型捕捉了数据集中存在的不同现象。

英文摘要:

The web provides a rich, open-domain environment with textual, structural, and spatial properties. We propose a new task for grounding language in this environment: given a natural language command (e.g., "click on the second article"), choose the correct element on the web page (e.g., a hyperlink or text box). We collected a dataset of over 50,000 commands that capture various phenomena such as functional references (e.g. "find who made this site"), relational reasoning (e.g. "article by john"), and visual reasoning (e.g. "top-most article"). We also implemented and analyzed three baseline models that capture different phenomena present in the dataset.

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。

↑