Taming CATS: Controllable Automatic Text Simplification through Instruction Fine-Tuning with Control Tokens
This work addresses the challenge of producing user-tailored simplified text for domains like medicine and news, though it is incremental as it builds on existing methods with a focus on data and evaluation limitations.
The paper tackled the problem of controllable automatic text simplification by introducing an instruction fine-tuning framework with control tokens, finding that smaller models (1-3B) can be competitive but controllability depends on data variation, with readability control learned consistently while compression control underperforms due to limited signal in existing corpora.
Controllable Automatic Text Simplification (CATS) produces user-tailored outputs, yet controllability is often treated as a decoding problem and evaluated with metrics that are not reflective to the measure of control. We observe that controllability in ATS is significantly constrained by data and evaluation. To this end, we introduce a domain-agnostic CATS framework based on instruction fine-tuning with discrete control tokens, steering open-source models to target readability levels and compression rates. Across three model families with different model sizes (Llama, Mistral, Qwen; 1-14B) and four domains (medicine, public administration, news, encyclopedic text), we find that smaller models (1-3B) can be competitive, but reliable controllability strongly depends on whether the training data encodes sufficient variation in the target attribute. Readability control (FKGL, ARI, Dale-Chall) is learned consistently, whereas compression control underperforms due to limited signal variability in the existing corpora. We further show that standard simplification and similarity metrics are insufficient for measuring control, motivating error-based measures for target-output alignment. Finally, our sampling and stratification experiments demonstrate that naive splits can introduce distributional mismatch that undermines both training and evaluation.