thunlp / SDLM-pytorch

Code accompanying EMNLP 2018 paper Language Modeling with Sparse Product of Sememe Experts

SDLM-pytorch

This repo contains codes accompanying paper Language Modeling with Sparse Product of Sememe Experts (EMNLP 2018): https://arxiv.org/abs/1810.12387.

The two folders are for two main experiments in the paper: Language Modeling and Headline Generation respectively. See readme files in those two folders for detailed usage.

Note: We have added an erratum and updated the results for the Headline Generation task. See readme file in the corresponding folder.

Any inquiry about bugs, feature request or questions about the paper are welcomed. You can directly post in Issues, or email any of us via:

• Hao Zhu: [e-mail]
• Yihong Gu: [e-mail]
• Jun Yan: [e-mail]

We will respond as soon as we can.