MAPLE: A Meta-learning Framework for Cross-Prompt Essay Scoring
This work addresses the cross-prompt generalization problem in automated essay scoring, a key bottleneck for real-world deployment.
MAPLE introduces a meta-learning framework using prototypical networks to improve automated essay scoring across unseen prompts, achieving state-of-the-art results on ELLIPSE and LAILA datasets with QWK improvements of 8.5 and 3 points, respectively.
Automated Essay Scoring (AES) faces significant challenges in cross-prompt settings, where models must generalize to unseen writing prompts. To address this limitation, we propose MAPLE, a meta-learning framework that leverages prototypical networks to learn transferable representations across different writing prompts. Across three diverse datasets (ELLIPSE and ASAP (English), and LAILA (Arabic)), MAPLE achieves state-of-the-art performance on ELLIPSE and LAILA, outperforming strong baselines by 8.5 and 3 points in QWK, respectively. On ASAP, where prompts exhibit heterogeneous score ranges, MAPLE yields improvements on several traits, highlighting the strengths of our approach in unified scoring settings. Overall, our results demonstrate the potential of meta-learning for building robust cross-prompt AES systems.