Texygen: A Benchmarking Platform for Text Generation Models
This provides a standardized tool for researchers working on text generation models, though it is incremental as it builds on existing models and metrics.
The authors introduced Texygen, a benchmarking platform for open-domain text generation models that implements various models and evaluation metrics for diversity, quality, and consistency, aiming to standardize research and improve reproducibility in the field.
We introduce Texygen, a benchmarking platform to support research on open-domain text generation models. Texygen has not only implemented a majority of text generation models, but also covered a set of metrics that evaluate the diversity, the quality and the consistency of the generated texts. The Texygen platform could help standardize the research on text generation and facilitate the sharing of fine-tuned open-source implementations among researchers for their work. As a consequence, this would help in improving the reproductivity and reliability of future research work in text generation.