Overcoming Conflicting Data when Updating a Neural Semantic Parser

David Gaddy, Alex Kouzemtchenko, Pavankumar Reddy Muddireddy, Prateek Kolhar, Rushin Shah


Abstract
In this paper, we explore how to use a small amount of new data to update a task-oriented semantic parsing model when the desired output for some examples has changed. When making updates in this way, one potential problem that arises is the presence of conflicting data, or out-of-date labels in the original training set. To evaluate the impact of this understudied problem, we propose an experimental setup for simulating changes to a neural semantic parser. We show that the presence of conflicting data greatly hinders learning of an update, then explore several methods to mitigate its effect. Our multi-task and data selection methods lead to large improvements in model accuracy compared to a naive data-mixing strategy, and our best method closes 86% of the accuracy gap between this baseline and an oracle upper bound.
Anthology ID:
2021.nlp4convai-1.5
Volume:
Proceedings of the 3rd Workshop on Natural Language Processing for Conversational AI
Month:
November
Year:
2021
Address:
Online
Venues:
EMNLP | NLP4ConvAI
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
40–51
Language:
URL:
https://aclanthology.org/2021.nlp4convai-1.5
DOI:
10.18653/v1/2021.nlp4convai-1.5
Bibkey:
Cite (ACL):
David Gaddy, Alex Kouzemtchenko, Pavankumar Reddy Muddireddy, Prateek Kolhar, and Rushin Shah. 2021. Overcoming Conflicting Data when Updating a Neural Semantic Parser. In Proceedings of the 3rd Workshop on Natural Language Processing for Conversational AI, pages 40–51, Online. Association for Computational Linguistics.
Cite (Informal):
Overcoming Conflicting Data when Updating a Neural Semantic Parser (Gaddy et al., NLP4ConvAI 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.nlp4convai-1.5.pdf
Code
 google/overcoming-conflicting-data