Metraipolis › archive › dc_0ea1206221cc

Error-Supervised Synthetic Learner Writing for Automated Essay Scoring — arXiv:2609.23573

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.23573
License (as archived)
CC0-1.0
Archived
2026-10-09T06:49:51.796Z
SHA-256 of the archived text
61b68422f320796a2fa0071e37b13754c28c6789906ca6412d66073666485e7c
Stable URL of this page
https://metraipolis.com/documento/dc_0ea1206221cc

How to cite

Error-Supervised Synthetic Learner Writing for Automated Essay Scoring — arXiv:2609.23573 — arXiv (en), https://arxiv.org/abs/2609.23573 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0ea1206221cc (Metraipolis archive dc_0ea1206221cc, sha256 61b68422f320796a2fa0071e37b13754c28c6789906ca6412d66073666485e7c)
BibTeX
@misc{dc0ea1206221cc,
  title = {Error-Supervised Synthetic Learner Writing for Automated Essay Scoring — arXiv:2609.23573},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.23573},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0ea1206221cc; sha256 61b68422f320796a2fa0071e37b13754c28c6789906ca6412d66073666485e7c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23573. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Synthetic essays can help reduce dependence on human-written data in Automated Essay Scoring (AES). However, they often lack realistic errors, limiting their ability to represent authentic human writing, particularly when the target texts are intended to resemble those produced by language learners. In this study, we present a simple approach that introduces error supervision into synthetic essay generation. Specifically, we fine-tune an LLM generator on error-annotated texts of the kind commonly used in Grammatical Error Detection (GED). To assess the utility of the proposed approach, we fine-tune and evaluate AES scorers under three data conditions: authentic essays, synthetic essays generated conventionally, and synthetic essays generated using our proposed approach. The results show that in the larger-data settings, the proposed approach outperforms the conventional synthetic baseline in 11 out of 12 dataset-metric comparisons, with performance in some cases approaching that of models trained on authentic essays. Despite these gains, performance under extremely low-resource settings remains mixed, with advantages over the conventional baseline only becoming more apparent at 200 training essays, although not consistently across datasets. Qualitative and quantitative analyses further show that the proposed approach produces learner-like errors whose distributions broadly resemble those observed in authentic essays.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.23573 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T06:48:44.996Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).