Abstract
We propose to tackle data-to-text generation tasks by directly splicing together retrieved segments of text from “neighbor” source-target pairs. Unlike recent work that conditions on retrieved neighbors but generates text token-by-token, left-to-right, we learn a policy that directly manipulates segments of neighbor text, by inserting or replacing them in partially constructed generations. Standard techniques for training such a policy require an oracle derivation for each generation, and we prove that finding the shortest such derivation can be reduced to parsing under a particular weighted context-free grammar. We find that policies learned in this way perform on par with strong baselines in terms of automatic and human evaluation, but allow for more interpretable and controllable generation.- Anthology ID:
- 2021.emnlp-main.352
- Volume:
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Month:
- November
- Year:
- 2021
- Address:
- Online and Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4283–4299
- Language:
- URL:
- https://aclanthology.org/2021.emnlp-main.352
- DOI:
- 10.18653/v1/2021.emnlp-main.352
- Cite (ACL):
- Sam Wiseman, Arturs Backurs, and Karl Stratos. 2021. Data-to-text Generation by Splicing Together Nearest Neighbors. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 4283–4299, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Data-to-text Generation by Splicing Together Nearest Neighbors (Wiseman et al., EMNLP 2021)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-3/2021.emnlp-main.352.pdf
- Code
- swiseman/neighbor-splicing
- Data
- WikiBio