Retrieval-augmented Generation for Automated Written Corrective Feedback: From Dataset to Human Evaluation

Jasper Degraeuwe; Thomas Moerman

Retrieval-augmented Generation for Automated Written Corrective Feedback: From Dataset to Human Evaluation

CLIN 35, 2025

Educational NLP

Retrieval-augmented generation for automated written corrective feedback in second language acquisition.

Authors

Jasper Degraeuwe

Thomas Moerman

Published

June 1, 2025

Work on retrieval-augmented generation for automated written corrective feedback in second language acquisition, moving from dataset construction toward human evaluation.

Research theme: Educational NLP

Citation

BibTeX citation:

@inproceedings{degraeuwe2025,
  author = {Degraeuwe, Jasper and Moerman, Thomas},
  title = {Retrieval-Augmented {Generation} for {Automated} {Written}
    {Corrective} {Feedback:} {From} {Dataset} to {Human} {Evaluation}},
  booktitle = {35th Meeting of Computational Linguistics in the
    Netherlands (CLIN 35)},
  date = {2025-06-01},
  langid = {en}
}

For attribution, please cite this work as:

Degraeuwe, Jasper, and Thomas Moerman. 2025. “Retrieval-Augmented Generation for Automated Written Corrective Feedback: From Dataset to Human Evaluation.” 35th Meeting of Computational Linguistics in the Netherlands (CLIN 35), accepted, June 1.