arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.14899cs.CL
收藏

Generate-and-Encoding:使用预测来改进语义解析的检索

Generate-and-Retrieve: use your predictions to improve retrieval for semantic parsing

University of Southern California(南加州大学);Google Research(谷歌研究院)

Yury Zemlyanskiy, Michiel de Jong, Joshua Ainslie, Panupong Pasupat, Peter Shaw, Linlu Qiu, Sumit Sanghai, Fei Sha

摘要:

最近一种常见的语义解析方法通过检索和附加一组训练样本(称为样本)来增强序列到序列模型。这个配方的有效性受到检索信息样本的能力的限制,这些样本有助于产生正确的解析,这在资源匮乏的环境中尤其具有挑战性。现有的检索通常是基于相似性的查询和范例输入。我们提出GandR,检索过程,检索样本的输出也是类似的。GandRfirst通过基于输入的检索生成初步预测。然后,它检索输出类似于用于生成最终预测的初步预测的样本。GandR在多个低资源语义解析任务上设置了最先进的技术。

英文摘要:

A common recent approach to semantic parsing augments sequence-to-sequence models by retrieving and appending a set of training samples, called exemplars. The effectiveness of this recipe is limited by the ability to retrieve informative exemplars that help produce the correct parse, which is especially challenging in low-resource settings. Existing retrieval is commonly based on similarity of query and exemplar inputs. We propose GandR, a retrieval procedure that retrieves exemplars for which outputs are also similar. GandRfirst generates a preliminary prediction with input-based retrieval. Then, it retrieves exemplars with outputs similar to the preliminary prediction which are used to generate a final prediction. GandR sets the state of the art on multiple low-resource semantic parsing tasks.

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。

↑