Simple and Effective Retrieve-Edit-Rerank Text Generation

Nabil Hossain, Marjan Ghazvininejad, Luke Zettlemoyer


Abstract
Retrieve-and-edit seq2seq methods typically retrieve an output from the training set and learn a model to edit it to produce the final output. We propose to extend this framework with a simple and effective post-generation ranking approach. Our framework (i) retrieves several potentially relevant outputs for each input, (ii) edits each candidate independently, and (iii) re-ranks the edited candidates to select the final output. We use a standard editing model with simple task-specific re-ranking approaches, and we show empirically that this approach outperforms existing, significantly more complex methodologies. Experiments on two machine translation (MT) datasets show new state-of-art results. We also achieve near state-of-art performance on the Gigaword summarization dataset, where our analyses show that there is significant room for performance improvement with better candidate output selection in future work.
Anthology ID:
2020.acl-main.228
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2532–2538
Language:
URL:
https://aclanthology.org/2020.acl-main.228
DOI:
10.18653/v1/2020.acl-main.228
Bibkey:
Cite (ACL):
Nabil Hossain, Marjan Ghazvininejad, and Luke Zettlemoyer. 2020. Simple and Effective Retrieve-Edit-Rerank Text Generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 2532–2538, Online. Association for Computational Linguistics.
Cite (Informal):
Simple and Effective Retrieve-Edit-Rerank Text Generation (Hossain et al., ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.228.pdf
Video:
 http://slideslive.com/38929289