Active Grade Estimator (AGE)

Research Method · All research methods
Evidence · 3 claims (3 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 3 claims rest on one study

Description

AGE is a pool-based active learning model for automated grading systems (AGSs) that integrates LLMs and regression models. It consists of three major modules: "two-stage sampling, assessment voting, and grading estimation modules". The two-stage query first selects the most contextually similar student answer via cosine similarity, then rewrites it via LLM canonicalization and simplification before teacher annotation. A voting mechanism based on minimum error grading estimation reduces annotation bias from multiple teachers, and a regression model estimates grades for remaining ungraded answers.

Accounts

• Active Grade Estimator (AGE): a pool-based active learning model for automated short answer grading with two-stage sampling, assessment voting, and grading estimation modules: AGE is a pool-based active learning model for automated grading systems (AGSs) that integrates LLMs and regression models. It consists of three major modules: "two-stage sampling, assessment voting, and grading estimation modules". The two-stage query first selects the most contextually similar student answer via cosine similarity, then rewrites it via LLM canonicalization and simplification before teacher annotation. A voting mechanism based on minimum error grading estimation reduces annotation bias from multiple teachers, and a regression model estimates grades for remaining ungraded answers. (Mohammad Iqbal et al. (2024))

Claims

• On the mismatch-grade dataset, all models without the voting mechanism diverge, while AGE with VM estimates student answers increasingly precisely [+M]
• AGE reduces teacher grading burden by requiring only 1% of student answers in the UNT dataset to be graded [+W]
• Instead of cutting answers, the proposed model converges faster than conventional AL and elevates grading up to 30% at the very beginning on the JRL dataset [+W]

Related Research Methods

-

Key Sources

• Mohammad Iqbal, Tsamara Rana Nugraha and Hsing-Kuo Pao. (2024). Active Grade Estimator on Short Answer Assessment. https://doi.org/10.21203/rs.3.rs-5651820/v1

Page metadata

Field	Value
type	research-method
id	active-grade-estimator-age
title	Active Grade Estimator (AGE)
description	AGE is a pool-based active learning model for automated grading systems (AGSs) that integrates LLMs and regression models.
status	review
generated	by process:settle-candidates on 2026-10-28
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.