Keep it Simple: Unsupervised Simplification of Multi-Paragraph Text
This work addresses the challenge of simplifying complex texts for improved readability and comprehension, particularly in news domains, representing a novel unsupervised approach with practical gains.
The authors tackled the problem of unsupervised multi-paragraph text simplification by developing the Keep it Simple (KiS) model, which balances fluency, salience, and simplicity using a novel reward optimization algorithm, resulting in a 4+ SARI point improvement over supervised baselines and enabling 18% faster text comprehension with retained accuracy.
This work presents Keep it Simple (KiS), a new approach to unsupervised text simplification which learns to balance a reward across three properties: fluency, salience and simplicity. We train the model with a novel algorithm to optimize the reward (k-SCST), in which the model proposes several candidate simplifications, computes each candidate's reward, and encourages candidates that outperform the mean reward. Finally, we propose a realistic text comprehension task as an evaluation method for text simplification. When tested on the English news domain, the KiS model outperforms strong supervised baselines by more than 4 SARI points, and can help people complete a comprehension task an average of 18% faster while retaining accuracy, when compared to the original text. Code available: https://github.com/tingofurro/keep_it_simple