Kevin Gu, Eva Tuecke, et al.
ICML 2024
Models for text generation have become focal for many research tasks and especially for the generation of sentence corpora. However, understanding the properties of an automatically generated text corpus remains challenging. We propose a set of tools that examine the properties of generated text corpora. Applying these tools on various generated corpora allowed us to gain new insights into the properties of the generative models. As part of our characterization process, we found remarkable differences in the corpora generated by two leading generative technologies.
Kevin Gu, Eva Tuecke, et al.
ICML 2024
Felipe Maia Polo, Lucas Weber, et al.
ICLR 2024
Sumit Neelam, Udit Sharma, et al.
EMNLP 2022
Shashanka Ubaru, Sanjeeb Dash, et al.
NeurIPS 2020