The 8088 The 8088 ← All news
arXiv cs.AI AI Research Apr 22

Has Automated Essay Scoring Reached Sufficient Accuracy? Deriving Achievable QWK Ceilings from Classical Test Theory

★★★★★ significance 2/5

The paper investigates the theoretical limits of Automated Essay Scoring (AES) accuracy using quadratic weighted kappa (QWK). It derives two specific ceilings—theoretical and human-like—to better understand the gap between current model performance and achievable accuracy under label noise.

Why it matters Establishes the theoretical upper bounds of model performance to distinguish between genuine algorithmic progress and the inherent noise of human grading.
Read the original at arXiv cs.AI

Tags

#automated essay scoring #quadratic weighted kappa #classical test theory #model evaluation

Related coverage