TY - GEN
T1 - Investigating Paraphrase Generation as a Data Augmentation Strategy for Low-Resource AMR-to-Text Generation
AU - Cabezudo, Marco Antonio Sobrevilla
AU - Inácio, Marcio Lima
AU - Pardo, Thiago Alexandre Salgueiro
N1 - Publisher Copyright:
© 2024 Association for Computational Linguistics.
PY - 2024
Y1 - 2024
N2 - Meaning Representation (AMR) is a meaning representation (MR) designed to abstract away from syntax, allowing syntactically different sentences to share the same AMR graph. Unlike other MRs, existing AMR corpora typically link one AMR graph to a single reference. This paper investigates the value of paraphrase generation in low-resource AMR-to-Text generation by testing various paraphrase generation strategies and evaluating their impact. The findings show that paraphrase generation significantly outperforms the baseline and traditional data augmentation methods, even with fewer training instances. Human evaluations indicate that this strategy often produces syntactic-based paraphrases and can exceed the performance of previous approaches. Additionally, the paper releases a paraphrase-extended version of the AMR corpus.
AB - Meaning Representation (AMR) is a meaning representation (MR) designed to abstract away from syntax, allowing syntactically different sentences to share the same AMR graph. Unlike other MRs, existing AMR corpora typically link one AMR graph to a single reference. This paper investigates the value of paraphrase generation in low-resource AMR-to-Text generation by testing various paraphrase generation strategies and evaluating their impact. The findings show that paraphrase generation significantly outperforms the baseline and traditional data augmentation methods, even with fewer training instances. Human evaluations indicate that this strategy often produces syntactic-based paraphrases and can exceed the performance of previous approaches. Additionally, the paper releases a paraphrase-extended version of the AMR corpus.
UR - https://www.scopus.com/pages/publications/105016579413
M3 - Conference contribution
AN - SCOPUS:105016579413
T3 - INLG 2024 - 17th International Natural Language Generation Conference, Proceedings of the Conference
SP - 663
EP - 675
BT - INLG 2024 - 17th International Natural Language Generation Conference, Proceedings of the Conference
A2 - Mahamood, Saad
A2 - Minh, Nguyen Le
A2 - Ippolito, Daphne
PB - Association for Computational Linguistics (ACL)
T2 - 17th International Natural Language Generation Conference, INLG 2024
Y2 - 23 September 2024 through 27 September 2024
ER -