Abstract
Graph-to-text (G2T) generation takes a graph as input and aims to generate a fluent and faith- ful textual representation of the information in the graph. The task has many applications, such as dialogue generation and question an- swering. In this work, we investigate to what extent the G2T generation problem is solved for previously studied datasets, and how pro- posed metrics perform when comparing generated texts. To help address their limitations, we propose a new metric that correctly identifies factual faithfulness, i.e., given a triple (subject, predicate, object), it decides if the triple is present in a generated text. We show that our metric FactSpotter achieves the highest correlation with human annotations on data correct- ness, data coverage, and relevance. In addition, FactSpotter can be used as a plug-in feature to improve the factual faithfulness of existing models. Finally, we investigate if existing G2T datasets are still challenging for state-of-the-art models. Our code is available online: https://github.com/guihuzhang/FactSpotter.- Anthology ID:
- 2023.findings-emnlp.672
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 10025–10042
- Language:
- URL:
- https://aclanthology.org/2023.findings-emnlp.672
- DOI:
- 10.18653/v1/2023.findings-emnlp.672
- Cite (ACL):
- Kun Zhang, Oana Balalau, and Ioana Manolescu. 2023. FactSpotter: Evaluating the Factual Faithfulness of Graph-to-Text Generation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 10025–10042, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- FactSpotter: Evaluating the Factual Faithfulness of Graph-to-Text Generation (Zhang et al., Findings 2023)
- PDF:
- https://preview.aclanthology.org/aacl-23-doi-ingestion/2023.findings-emnlp.672.pdf