NL-EDIT: Correcting semantic parse errors through natural language interaction

We study semantic parsing in an interactive setting in which users correct\nerrors with natural language feedback. We present NL-EDIT, a model for\ninterpreting natural language feedback in the interaction context to generate a\nsequence of edits that can be applied to the initial parse to correct its\nerrors. We show that NL-EDIT can boost the accuracy of existing text-to-SQL\nparsers by up to 20% with only one turn of correction. We analyze the\nlimitations of the model and discuss directions for improvement and evaluation.\nThe code and datasets used in this paper are publicly available at\nhttp://aka.ms/NLEdit.\n

cs.CL Natural Language Processing Techniques Topic Modeling Software Engineering Research Computer Science

Paper

References (44)

01DuoRAT: Towards Simpler Text-to-SQL ModelsTorsten Scholak, Raymond Li, Dzmitry Bahdanau et al.2020 · North American Chapter of the Association for Computational Linguistics · 28 citations In Library

02AutoQA: From Databases to Q&A Semantic Parsers with Only Synthetic Training DataSilei Xu, Sina J. Semnani, Giovanni Campagna et al.2020 · Conference on Empirical Methods in Natural Language Processing · 55 citations In Library

03GraPPa: Grammar-Augmented Pre-Training for Table Semantic ParsingTao Yu, Chien-Sheng Wu, Xi Victoria Lin et al.2020 · International Conference on Learning Representations · 297 citations In Library

04Seq2Edits: Sequence Transduction Using Span-level Edit OperationsFelix Stahlberg, Shankar Kumar2020 · Conference on Empirical Methods in Natural Language Processing · 103 citations In Library

05Task-Oriented Dialogue as Dataflow SynthesisJacob Andreas, J. Bufe, David Burkett et al.2020 · Transactions of the Association for Computational Linguistics · 177 citations In Library

06TaBERT: Pretraining for Joint Understanding of Textual and Tabular DataPengcheng Yin, Graham Neubig, Wen-tau Yih et al.2020 · Annual Meeting of the Association for Computational Linguistics · 819 citations In Library

07Learning to Update Natural Language Comments Based on Code ChangesSheena Panthaplackel, Pengyu Nie, Miloš Gligorić et al.2020 · Annual Meeting of the Association for Computational Linguistics · 84 citations In Library

08Speak to your Parser: Interactive Text-to-SQL with Natural Language FeedbackAhmed Elgohary, Saghar Hosseini, Ahmed Hassan Awadallah2020 · Annual Meeting of the Association for Computational Linguistics · 81 citations In Library

09RYANSQL: Recursively Applying Sketch-based Slot Fillings for Complex Text-to-SQL in Cross-Domain DatabasesDonghyun Choi, M. Shin, EungGyun Kim et al.2020 · International Conference on Computational Logic · 161 citations In Library

10TaPas: Weakly Supervised Table Parsing via Pre-trainingJonathan Herzig, Pawel Krzysztof Nowak, Thomas Müller et al.2020 · Annual Meeting of the Association for Computational Linguistics · 894 citations In Library

11RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL ParsersBailin Wang, Richard Shin, Xiaodong Liu et al.2019 · Annual Meeting of the Association for Computational Linguistics · 869 citations In Library

12Model-based Interactive Semantic Parsing: A Unified Framework and A Text-to-SQL Case StudyZiyu Yao, Yu Su, Huan Sun et al.2019 · Conference on Empirical Methods in Natural Language Processing · 89 citations In Library

Scroll for more · 32 remaining

Similar papers

NL-EDIT: Correcting Semantic Parse Errors through Natural Language InteractionAhmed Elgohary, Christopher Meek, Matthew Richardson et al.2021 In Library

Speak to your Parser: Interactive Text-to-SQL with Natural Language FeedbackAhmed Elgohary, Saghar Hosseini, Ahmed Hassan Awadallah2020 · arXiv (Cornell University) · 81 citations In Library

Speak to your Parser: Interactive Text-to-SQL with Natural Language FeedbackAhmed Elgohary, Saghar Hosseini, Ahmed Hassan Awadallah2020 In Library

Learning to Simulate Natural Language Feedback for Interactive Semantic ParsingHao Yan, Saurabh Srivastava, Yintao Tai et al.2023 In Library

Learning to Simulate Natural Language Feedback for Interactive Semantic ParsingHao Yan, Saurabh Srivastava, Yintao Tai et al.2023 · arXiv (Cornell University) · 25 citations In Library

Correcting Semantic Parses with Natural Language through Dynamic Schema EncodingParker Glenn, Parag Pravin Dakle, Preethi Raghavan2023 In Library

Correcting Semantic Parses with Natural Language through Dynamic Schema EncodingParker Glenn, Parag Pravin Dakle, Preethi Raghavan2023 · arXiv (Cornell University) In Library

Learning a Neural Semantic Parser from User FeedbackSrini Iyer, Ioannis Konstas, Alvin Cheung et al.2017 · Annual Meeting of the Association for Computational Linguistics · 377 citations In Library

Interactively Diagnosing Errors in a Semantic ParserConstantine Nakos, Kenneth D. Forbus2024 · arXiv (Cornell University) In Library

Text-to-SQL Error Correction with Language Models of CodeZiru Chen, Shijie Chen, Michael White et al.2023 · arXiv (Cornell University) · 34 citations In Library

Source of record

arXiv

44B

Welcome.

This panel sticks with you. Pick anything below and it opens right beside it, so you can dig through 60,000-plus records without ever losing your spot here.

Travel 44B

LibraryPapers, policy, standards, statute LabsEvery organization building AI ModelsIntelligence, price, and speed BenchmarksThe evaluation catalog IncidentsAI failures and harms SearchOne field across everything DashboardThe state of AI accountability in NY 44B RegistryThe Art. 44-B compliance portal