Abstract
Abstract We introduce an Edit-Based TransfOrmer with Repositioning (EDITOR), which makes sequence generation flexible by seamlessly allowing users to specify preferences in output lexical choice. Building on recent models for non-autoregressive sequence generation (Gu et al., 2019), EDITOR generates new sequences by iteratively editing hypotheses. It relies on a novel reposition operation designed to disentangle lexical choice from word positioning decisions, while enabling efficient oracles for imitation learning and parallel edits at decoding time. Empirically, EDITOR uses soft lexical constraints more effectively than the Levenshtein Transformer (Gu et al., 2019) while speeding up decoding dramatically compared to constrained beam search (Post and Vilar, 2018). EDITOR also achieves comparable or better translation quality with faster decoding speed than the Levenshtein Transformer on standard Romanian-English, English-German, and English-Japanese machine translation tasks.- Anthology ID:
- 2021.tacl-1.19
- Volume:
- Transactions of the Association for Computational Linguistics, Volume 9
- Month:
- Year:
- 2021
- Address:
- Cambridge, MA
- Venue:
- TACL
- SIG:
- Publisher:
- MIT Press
- Note:
- Pages:
- 311–328
- Language:
- URL:
- https://aclanthology.org/2021.tacl-1.19
- DOI:
- 10.1162/tacl_a_00368
- Cite (ACL):
- Weijia Xu and Marine Carpuat. 2021. EDITOR: An Edit-Based Transformer with Repositioning for Neural Machine Translation with Soft Lexical Constraints. Transactions of the Association for Computational Linguistics, 9:311–328.
- Cite (Informal):
- EDITOR: An Edit-Based Transformer with Repositioning for Neural Machine Translation with Soft Lexical Constraints (Xu & Carpuat, TACL 2021)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2021.tacl-1.19.pdf