TextBox: A Unified, Modularized, and Extensible Framework for Text Generation
This work provides a comprehensive and modular framework for researchers and practitioners in natural language generation to reproduce baselines and develop new models.
This paper introduces TextBox, an open-source PyTorch library for text generation. It implements 21 models across 9 benchmark datasets, covering VAE, GAN, and pretrained language models, to offer a unified and modular framework for researchers.
In this paper, we release an open-source library, called TextBox, to provide a unified, modularized, and extensible text generation framework. TextBox aims to support a broad set of text generation tasks and models. In our library, we implement 21 text generation models on 9 benchmark datasets, covering the categories of VAE, GAN, and pretrained language models. Meanwhile, our library maintains sufficient modularity and extensibility by properly decomposing the model architecture, inference, and learning process into highly reusable modules, which allows users to easily incorporate new models into our framework. The above features make TextBox specially suitable for researchers and practitioners to quickly reproduce baseline models and develop new models. TextBox is implemented based on PyTorch, and released under Apache License 2.0 at https://github.com/RUCAIBox/TextBox.