الملخص
Grammatical error correction (GEC) evaluation has traditionally relied on reference or edit overlap, which can penalize valid rewrites that differ from gold corrections. Reference-free metrics reduce this dependence, but evaluating whether a fluent output is a valid correction of the source remains challenging. We propose SURE, a source-conditioned reward evaluator trained on within-source preferences spanning minimal-edit and rewrite-oriented corrections. SURE jointly learns an overall reward with criteria-level supervision for grammaticality, faithfulness, and fluency, together with span-level grounding for source-side error resolution. Experiments on SEEDA show that SURE performs competitively against strong baselines, with particular gains on rewrite-style corrections and more disentangled criteria-level diagnostics. Our code is available at https://github.com/hayeonggg/SURE.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Ryu, H., Jo, S., Yu, S., & Kim, Y. (2026). Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction. https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction
MLA 9
Ryu, Hayeong, et al. "Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction." https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction.
شيكاغو (المؤلف–التاريخ)
Ryu, Hayeong, Sunhee Jo, Seunguk Yu, and YoungBin Kim. 2026. "Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction." https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction.
هارفارد
Ryu, H., Jo, S., Yu, S. and Kim, Y. (2026) 'Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction', Available at: https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction.
فانكوفر
Ryu H, Jo S, Yu S, Kim Y. Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction. https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction
IEEE
H. Ryu, S. Jo, S. Yu, and Y. Kim, "Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction," https://omanscience.com/ar/articles/don-t-count-the-edits-judge-by-the-outcome-alone-reward-based-evaluation-for-grammatical-error-correction.