CLAILGNov 26, 2017

Improved Neural Text Attribute Transfer with Non-parallel Data

arXiv:1711.09395v239 citations
Originality Synthesis-oriented
AI Analysis

This work addresses text attribute transfer for natural language processing applications, but it appears incremental as it builds on existing approaches.

The paper tackled text attribute transfer with non-parallel data by proposing improvements to an encoder-decoder framework, resulting in outperforming a strong baseline in two out of three evaluation metrics on sentiment transfer tasks across two datasets.

Text attribute transfer using non-parallel data requires methods that can perform disentanglement of content and linguistic attributes. In this work, we propose multiple improvements over the existing approaches that enable the encoder-decoder framework to cope with the text attribute transfer from non-parallel data. We perform experiments on the sentiment transfer task using two datasets. For both datasets, our proposed method outperforms a strong baseline in two of the three employed evaluation metrics.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes