EMNLP Findings2020
Multichannel Generative Language Model: Learning All Possible Factorizations Within and Across Channels
Abstract
A channel corresponds to a viewpoint or transformation of an underlying meaning. A pair of parallel sentences in English and French express the same underlying meaning, but through two separate channels corresponding to their languages. In this work, we present the Multichannel Generative Language Model (MGLM). MGLM is a generative joint distribution model over channels. MGLM marginalizes over all possible factorizations within and across all channels. MGLM endows flexible inference, including unconditional generation, conditional generation (where 1 channel is observed and other channels are generated), and partially observed generation (where incomplete observations are spread across all the channels). We experiment with the Multi30K dataset containing English, French, Czech, and German. We demonstrate experiments with unconditional, conditional, and partially conditional generation. We provide qualitative samples sampled unconditionally from the generative joint distribution. We also quantitatively analyze the quality-diversity trade-offs and find MGLM outperforms traditional bilingual discriminative models.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on1
Related papers
- Improving Multilingual Neural Machine Translation with Auxiliary Source LanguagesWeijia Xu, Yuwei Yin, Shuming Ma, Dongdong Zhang et al.EMNLP Findings 2021 · 6 citations
- Mirror-Generative Neural Machine TranslationZaixiang Zheng, Hao Zhou, Shujian Huang, Lei Li et al.ICLR 2020 · 37 citations
- A Multilingual View of Unsupervised Machine TranslationXavier Garcia, Pierre Foret, Thibault Sellam, Ankur P. ParikhEMNLP Findings 2020 · 31 citations
- BERTGen: Multi-task Generation through BERTFaidon Mitzalis, Ozan Caglayan, Pranava Madhyastha, Lucia SpeciaACL 2021
- Bridging the Gap between Synthetic and Authentic Images for Multimodal Machine TranslationWenyu Guo, Qingkai Fang, Dong Yu, Yang FengEMNLP 2023 · 5 citations
