paper
arXiv cs.CL
November 4th, 2026 at 5:00 AMMitigating Bias in Automated Essay Scoring for ESL Learners via Contrastive Learning
arXiv:2601.16724v3 Announce Type: replace Abstract: Automated Essay Scoring systems disproportionately penalize high-proficiency English as a Second Language (ESL) learners. We propose Contrastive Learning with Matched Essay Pairs (CL-MEP), a bi-directional alignment strategy. CL-MEP reduces this scoring bias by 39.9% while improving overall accuracy, successfully disentangling valid syntactic complexity from surface-level grammatical errors.
#ai
Get stories like this every Monday
The week's most important AI news in one email. Free, unsubscribe anytime.
Score: 3.01
Engagement proxy: 0
Canonical link: https://arxiv.org/abs/2601.16724