Abstract
Neural text generation, including neural machine translation, image captioning, and summarization, has been quite successful recently. However, during training time, typically only one reference is considered for each example, even though there are often multiple references available, e.g., 4 references in NIST MT evaluations, and 5 references in image captioning data. We first investigate several different ways of utilizing multiple human references during training. But more importantly, we then propose an algorithm to generate exponentially many pseudo-references by first compressing existing human references into lattices and then traversing them to generate new pseudo-references. These approaches lead to substantial improvements over strong baselines in both machine translation (+1.5 BLEU) and image captioning (+3.1 BLEU / +11.7 CIDEr).- Anthology ID:
- D18-1357
- Volume:
- Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
- Month:
- October-November
- Year:
- 2018
- Address:
- Brussels, Belgium
- Venue:
- EMNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 3188–3197
- Language:
- URL:
- https://aclanthology.org/D18-1357
- DOI:
- 10.18653/v1/D18-1357
- Cite (ACL):
- Renjie Zheng, Mingbo Ma, and Liang Huang. 2018. Multi-Reference Training with Pseudo-References for Neural Translation and Text Generation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 3188–3197, Brussels, Belgium. Association for Computational Linguistics.
- Cite (Informal):
- Multi-Reference Training with Pseudo-References for Neural Translation and Text Generation (Zheng et al., EMNLP 2018)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/D18-1357.pdf
- Data
- COCO