Plan-then-Generate: Controlled Data-to-Text Generation via Planning
| cam.depositDate | 2022-05-27 | |
| cam.issuedOnline | 2021-11 | |
| cam.orpheus.counter | 11 | |
| cam.orpheus.success | 2022-08-30 | |
| dc.contributor.author | Su, Y | |
| dc.contributor.author | Vandyke, D | |
| dc.contributor.author | Wang, S | |
| dc.contributor.author | Fang, Y | |
| dc.contributor.author | Collier, N | |
| dc.contributor.orcid | Su, Yixuan [0000-0002-1472-7791] | |
| dc.contributor.orcid | Collier, Nigel [0000-0002-7230-4164] | |
| dc.date.accessioned | 2022-05-27T23:30:41Z | |
| dc.date.available | 2022-05-27T23:30:41Z | |
| dc.date.issued | 2021 | |
| dc.date.updated | 2022-05-27T06:37:38Z | |
| dc.description.abstract | Recent developments in neural networks have led to the advance in data-to-text generation. However, the lack of ability of neural models to control the structure of generated output can be limiting in certain real-world applications. In this study, we propose a novel Plan-then-Generate (PlanGen) framework to improve the controllability of neural data-totext models. Extensive experiments and analyses are conducted on two benchmark datasets, ToTTo and WebNLG. The results show that our model is able to control both the intrasentence and inter-sentence structure of the generated output. Furthermore, empirical comparisons against previous state-of-the-art methods show that our model improves the generation quality as well as the output diversity as judged by human and automatic evaluations. | |
| dc.identifier.doi | 10.17863/CAM.84984 | |
| dc.identifier.isbn | 9781955917100 | |
| dc.identifier.uri | https://www.repository.cam.ac.uk/handle/1810/337575 | |
| dc.language.iso | eng | |
| dc.publisher | Association for Computational Linguistics | |
| dc.publisher.department | Department of Theoretical & Applied Linguistics | |
| dc.publisher.department | Faculty of Modern And Medieval Languages And Linguistics | |
| dc.publisher.url | http://dx.doi.org/10.18653/v1/2021.findings-emnlp.76 | |
| dc.rights | Attribution 4.0 International | |
| dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
| dc.title | Plan-then-Generate: Controlled Data-to-Text Generation via Planning | |
| dc.type | Conference Object | |
| dcterms.dateAccepted | 2021-08-25 | |
| prism.endingPage | 909 | |
| prism.publicationDate | 2021 | |
| prism.publicationName | Findings of the Association for Computational Linguistics, Findings of ACL: EMNLP 2021 | |
| prism.startingPage | 895 | |
| pubs.conference-finish-date | 2021-11 | |
| pubs.conference-name | Findings of the Association for Computational Linguistics: EMNLP 2021 | |
| pubs.conference-start-date | 2021-11 | |
| pubs.licence-display-name | Apollo Repository Deposit Licence Agreement | |
| pubs.licence-identifier | apollo-deposit-licence-2-1 | |
| rioxxterms.version | VoR | |
| rioxxterms.versionofrecord | 10.18653/v1/2021.findings-emnlp.76 |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- plangen.pdf
- Size:
- 1.51 MB
- Format:
- Adobe Portable Document Format
- Description:
- Published version
- Licence
- https://creativecommons.org/licenses/by/4.0/