papers AI Learner
The Github is limit! Click to go to the new site.

Multi-Reference Training with Pseudo-References for Neural Translation and Text Generation

2018-08-28
Renjie Zheng, Mingbo Ma, Liang Huang

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).

Abstract (translated by Google)
URL

https://arxiv.org/abs/1808.09564

PDF

https://arxiv.org/pdf/1808.09564


Similar Posts

Comments