Probing Omissions and Distortions in Transformer-based RDF-to-Text Models
Transactions of the Association for Computational Linguistics · 05.11.2026 · 14 · 2757-2774
DOI: 10.1162/tacl.a.824 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 05.11.2026. Verfasst von Juliette Faille und 2 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
In Natural Language Generation (NLG), important information is sometimes omitted in the output text. To better understand and analyse how this type of mistake arises, we focus on RDF-to-Text generation and explore two methods of probing omissions in the encoder output of BART (Lewis et al., 2020) and of T5 (Raffel et al., 2019): (i) a novel parameter-free probing method based on the computation of cosine similarity between embeddings of RDF graphs and of RDF graphs in which we removed some entities and (ii) a parametric probe which performs binary classification on the encoder embeddings to detect omitted entities. We also extend our analysis to distorted entities, i.e. entities that are not fully correctly mentioned in the generated text (e.g. misspelling of entity, wrong units of measurement). We found that both omitted and distorted entities can be probed in the encoder’s output embeddings. This suggests that the encoder emits a weaker signal for these entities and therefore is responsible for some loss of information. This also shows that probing methods can be used to detect mistakes in the output of NLG models.
Zitieren
APA
Faille, J., Gatt, A., & Gardent, C. (2026). Probing Omissions and Distortions in Transformer-based RDF-to-Text Models. Transactions of the Association for Computational Linguistics, 14, 2757-2774. https://doi.org/10.1162/tacl.a.824
Vancouver
Faille J, Gatt A, Gardent C. Probing Omissions and Distortions in Transformer-based RDF-to-Text Models. Transactions of the Association for Computational Linguistics. 2026;14:2757-2774. doi:10.1162/tacl.a.824
BibTeX
@article{Faille2026,
author = {Faille, Juliette and Gatt, Albert and Gardent, Claire},
title = {Probing Omissions and Distortions in Transformer-based RDF-to-Text Models},
journal = {Transactions of the Association for Computational Linguistics},
year = {2026},
volume = {14},
pages = {2757--2774},
doi = {10.1162/tacl.a.824}
}
EUVIMED