SEFORA: Student Essays with Feedback Corpus and LLM Feedback Evaluation Framework
For researchers and educators developing automated writing feedback systems, this work provides a much-needed corpus and evaluation method, but the results are incremental as they primarily highlight current limitations.
The paper introduces SEFORA, a public corpus of 564 student essay drafts with 8,240 instructor annotations, and UniMatch, a reference-based evaluation framework for LLM-generated feedback. UniMatch shows that no LLM configuration exceeds 0.4 F1, indicating models struggle to align with instructor feedback priorities.
Effective writing feedback is among the strongest drivers of student learning, yet producing it at scale is labor-intensive. LLMs offer a natural path to scaling writing support, but two gaps stand in the way: few public corpora capture how instructors actually deliver feedback in real classrooms, and no reliable method measures whether generated feedback aligns with what an instructor would write. We address both. SEFORA is a public corpus pairing instructor inline feedback with assignment prompts, rubrics, scores, and multi-draft revisions across various college writing genres, comprising 564 drafts and 8,240 instructor annotations. UniMatch is a reference-based evaluation framework for open-ended generation: it segments feedback into feedback units, scores their semantic correspondence under instructor-derived criteria, and aligns them via optimal matching to yield interpretable precision, recall, and F1. Across 74 experimental configurations spanning multiple LLMs, no setting exceeds 0.4 F1. UniMatch reveals that models struggle to identify the feedback instructors would prioritize, and performance degrades as models generate more.