Controlling Pre-trained Language Models for Grade-Specific Text Simplification
Submission Type: Regular Long Paper
Submission Track: Natural Language Generation
Keywords: Controllable Text generation, Text Simplification
TL;DR: We improve the controllability of text simplification based on large language models by predicting control tokens that specify the nature of TS operations needed for each input.
Abstract: Text simplification systems rewrite text to make it more readable while preserving its content. However, what makes a text easy to read depends on the intended readers. Recent work has shown that pre-trained language models can simplify text using a wealth of techniques to control output simplicity, ranging from specifying only the desired reading grade level, to directly specifying low-level edit operations. Yet it remains unclear how to set these control parameters in practice. Existing approaches set them at the corpus level, disregarding the complexity of individual inputs and considering only one level of output complexity. In this work, we conduct an empirical study to understand how different control mechanisms impact the adequacy and simplicity of text simplification systems. Based on these insights, we introduce a simple method that predicts the edit operations required for simplifying a text for a specific grade level on an instance-per-instance basis. This approach improves the quality of the simplified outputs over corpus-level search-based heuristics.
Submission Number: 1633
Loading