Unsupervised Editing for Counterfactual Stories

Authors

  • Jiangjie Chen Fudan University ByteDance AI Lab
  • Chun Gan JD.com
  • Sijie Cheng Fudan University
  • Hao Zhou ByteDance AI Lab
  • Yanghua Xiao Fudan University Fudan-Aishu Cognitive Intelligence Research Center
  • Lei Li University of California Santa Barbara

DOI:

https://doi.org/10.1609/aaai.v36i10.21290

Keywords:

Speech & Natural Language Processing (SNLP)

Abstract

Creating what-if stories requires reasoning about prior statements and possible outcomes of the changed conditions. One can easily generate coherent endings under new conditions, but it would be challenging for current systems to do it with minimal changes to the original story. Therefore, one major challenge is the trade-off between generating a logical story and rewriting with minimal-edits. In this paper, we propose EDUCAT, an editing-based unsupervised approach for counterfactual story rewriting. EDUCAT includes a target position detection strategy based on estimating causal effects of the what-if conditions, which keeps the causal invariant parts of the story. EDUCAT then generates the stories under fluency, coherence and minimal-edits constraints. We also propose a new metric to alleviate the shortcomings of current automatic metrics and better evaluate the trade-off. We evaluate EDUCAT on a public counterfactual story rewriting benchmark. Experiments show that EDUCAT achieves the best trade-off over unsupervised SOTA methods according to both automatic and human evaluation. The resources of EDUCAT are available at: https://github.com/jiangjiechen/EDUCAT.

Downloads

Published

2022-06-28

How to Cite

Chen, J., Gan, C., Cheng, S., Zhou, H., Xiao, Y., & Li, L. (2022). Unsupervised Editing for Counterfactual Stories. Proceedings of the AAAI Conference on Artificial Intelligence, 36(10), 10473-10481. https://doi.org/10.1609/aaai.v36i10.21290

Issue

Section

AAAI Technical Track on Speech and Natural Language Processing