Unsupervised Editing for Counterfactual Stories
arXiv:2112.05417 · doi:10.1609/aaai.v36i10.21290
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.
arXiv commentAccepted to AAAI 2022
References in corpus (14)
- BERTScore: Evaluating Text Generation with BERT
- XLNet: Generalized Autoregressive Pretraining for Language Understanding
- CTRL: A Conditional Transformer Language Model for Controllable Generation
- Learning the Difference that Makes a Difference with Counterfactually-Augmented Data
- Plug and Play Language Models: A Simple Approach to Controlled Text Generation
- A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Stories
- Deep Generative Models with Learnable Knowledge Constraints
- Lexically Constrained Decoding for Sequence Generation Using Grid Beam Search
- A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation
- CGMH: Constrained Sentence Generation by Metropolis-Hastings Sampling
- NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraints
- AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples
- Show Me How To Revise: Improving Lexically Constrained Sentence Generation with XLNet
- Summarizing Text on Any Aspects: A Knowledge-Informed Weakly-Supervised Approach