Controlling Pre-trained Language Models for Grade-Specific Text Simplification
This work addresses the challenge of customizing text simplification for specific reader levels, though it is incremental as it builds on existing control mechanisms.
The paper tackled the problem of setting control parameters for pre-trained language models in grade-specific text simplification, introducing a method that predicts edit operations per instance, which improved output quality over corpus-level heuristics.
Text simplification (TS) 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.