Scope-enhanced Compositional Semantic Parsing for DRT

Xiulin Yang; Jonas Groschwitz; Alexander Koller; Johan Bos

doi:10.18653/v1/2024.emnlp-main.1093

Scope-enhanced Compositional Semantic Parsing for DRT

Xiulin Yang, Jonas Groschwitz, Alexander Koller, Johan Bos

Computational Linguistics (CL)

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Discourse Representation Theory (DRT) distinguishes itself from other semantic representation frameworks by its ability to model complex semantic and discourse phenomena through structural nesting and variable binding. While seq2seq models hold the state of the art on DRT parsing, their accuracy degrades with the complexity of the sentence, and they sometimes struggle to produce well-formed DRT representations. We introduce the AMS parser, a compositional, neurosymbolic semantic parser for DRT. It rests on a novel mechanism for predicting quantifier scope. We show that the AMS parser reliably produces well-formed outputs and performs well on DRT parsing, especially on complex sentences.

Original language	English
Title of host publication	Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Editors	Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Publisher	Association for Computational Linguistics, ACL Anthology
Pages	19602-19616
Number of pages	15
DOIs	https://doi.org/10.18653/v1/2024.emnlp-main.1093
Publication status	Published - 2024

Access to Document

10.18653/v1/2024.emnlp-main.1093Licence: CC BY

Scope-enhanced Compositional Semantic Parsing for DRTFinal publisher's version, 1.25 MBLicence: CC BY

Handle.net

Persistent link

Cite this

Yang, X., Groschwitz, J., Koller, A., & Bos, J. (2024). Scope-enhanced Compositional Semantic Parsing for DRT. In Y. Al-Onaizan, M. Bansal, & Y.-N. Chen (Eds.), Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (pp. 19602-19616). Association for Computational Linguistics, ACL Anthology. https://doi.org/10.18653/v1/2024.emnlp-main.1093

@inproceedings{d9870fe7e1944a17934549c9268795a6,

title = "Scope-enhanced Compositional Semantic Parsing for DRT",

abstract = "Discourse Representation Theory (DRT) distinguishes itself from other semantic representation frameworks by its ability to model complex semantic and discourse phenomena through structural nesting and variable binding. While seq2seq models hold the state of the art on DRT parsing, their accuracy degrades with the complexity of the sentence, and they sometimes struggle to produce well-formed DRT representations. We introduce the AMS parser, a compositional, neurosymbolic semantic parser for DRT. It rests on a novel mechanism for predicting quantifier scope. We show that the AMS parser reliably produces well-formed outputs and performs well on DRT parsing, especially on complex sentences.",

author = "Xiulin Yang and Jonas Groschwitz and Alexander Koller and Johan Bos",

year = "2024",

doi = "10.18653/v1/2024.emnlp-main.1093",

language = "English",

pages = "19602--19616",

editor = "Yaser Al-Onaizan and Mohit Bansal and Yun-Nung Chen",

booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",

publisher = "Association for Computational Linguistics, ACL Anthology",

}

Scope-enhanced Compositional Semantic Parsing for DRT. / Yang, Xiulin; Groschwitz, Jonas; Koller, Alexander et al.
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. ed. / Yaser Al-Onaizan; Mohit Bansal; Yun-Nung Chen. Association for Computational Linguistics, ACL Anthology, 2024. p. 19602-19616.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Scope-enhanced Compositional Semantic Parsing for DRT

AU - Yang, Xiulin

AU - Groschwitz, Jonas

AU - Koller, Alexander

AU - Bos, Johan

PY - 2024

Y1 - 2024

N2 - Discourse Representation Theory (DRT) distinguishes itself from other semantic representation frameworks by its ability to model complex semantic and discourse phenomena through structural nesting and variable binding. While seq2seq models hold the state of the art on DRT parsing, their accuracy degrades with the complexity of the sentence, and they sometimes struggle to produce well-formed DRT representations. We introduce the AMS parser, a compositional, neurosymbolic semantic parser for DRT. It rests on a novel mechanism for predicting quantifier scope. We show that the AMS parser reliably produces well-formed outputs and performs well on DRT parsing, especially on complex sentences.

AB - Discourse Representation Theory (DRT) distinguishes itself from other semantic representation frameworks by its ability to model complex semantic and discourse phenomena through structural nesting and variable binding. While seq2seq models hold the state of the art on DRT parsing, their accuracy degrades with the complexity of the sentence, and they sometimes struggle to produce well-formed DRT representations. We introduce the AMS parser, a compositional, neurosymbolic semantic parser for DRT. It rests on a novel mechanism for predicting quantifier scope. We show that the AMS parser reliably produces well-formed outputs and performs well on DRT parsing, especially on complex sentences.

U2 - 10.18653/v1/2024.emnlp-main.1093

DO - 10.18653/v1/2024.emnlp-main.1093

M3 - Conference contribution

SP - 19602

EP - 19616

BT - Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing

A2 - Al-Onaizan, Yaser

A2 - Bansal, Mohit

A2 - Chen, Yun-Nung

PB - Association for Computational Linguistics, ACL Anthology

ER -

Scope-enhanced Compositional Semantic Parsing for DRT

Abstract

Access to Document

Handle.net

Fingerprint

Cite this