DETECT

This repository contains the datasets and metric umplementation from our paper:
DETECT: Determining Ease and Textual Clarity of German Text Simplifications

---

📂 Repository Structure

SimpEvalDE/

Contains the instructions and resources for generating SimpEvalDE,
a German-specific dataset for text simplification quality evaluation on which DETECT was trained and evaluated.
SimpEvalDE includes synthetic scores for simplicity, meaning preservation,
and fluency, generated by three LLM-based judges.

lens/

Implements:

• LENS [Maddela et al., 2023]
• DETECT, our German-specific adaptation of LENS

Please refer to lens/README.md for detailed usage instructions,
model loading, and scoring examples.

📚 Citation

If you use DETECT in your work, please cite:

@inproceedings{korobeynikova2026detect,
  title     = {DETECT: Determining Ease and Textual Clarity of German Text Simplifications},
  author    = {Korobeynikova, Maria and Battisti, Alessia and Fischer, Lukas and Gao, Yingqiang},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages     = {2852--2882},
  year      = {2026}
}