Skip to content

PrairieLearn-integrated AI grading tool for handwritten mathematics with single-call transcription and rubric evaluation

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The article presents a prototype AI-assisted grading tool integrated into PrairieLearn, an open-source online assessment platform, that evaluates open-ended mathematical responses submitted as handwritten photographs. It extends a prior typed-response pipeline by introducing an image-based grading pipeline and prompt structure that jointly perform transcription and rubric-based evaluation. The prompt includes the problem statement, rubric, optional reference solutions, and explicit directives for both transcription and assessment. Human graders may review and override AI evaluations, and human-assigned grades always take precedence.

Design Implications

Context

Requirements

  • The LLM prompt includes the problem statement, rubric, optional reference solutions and instructions, and explicit directives for both transcription and assessment
  • The grader operates asynchronously, executing only after submissions close and the grading configuration is finalized

Constraints

  • Only the final uploaded image is graded
  • Randomized problem instances preclude retrieval-augmented grading, as the variability across problem instances prevents reliable reuse of previously graded examples

Target Learners

  • university STEM students in numerical methods and introductory dynamics courses

Learning Goals

  • grading of multi-step handwritten mathematical derivations against instructor-defined rubrics

Claims

Examples

-

Key Sources

  • Jacob Levine, Miguel Aenlle, Craig Zilles, Matthew West, and Mariana Silva. (2026). Automated Grading of Handwritten Mathematics Using Vision-Capable LLMs. https://arxiv.org/abs/2605.19043
Page metadata
Field Value
type design
id prairielearn-vision-llm-handwritten-math-grader
title PrairieLearn-integrated AI grading tool for handwritten mathematics with single-call transcription and rubric evaluation
description The article presents a prototype AI-assisted grading tool integrated into PrairieLearn, an open-source online assessment platform, that evaluates open-ended mathematical responses submitted as handwritten photographs.
status review
generated by process:settle-candidates on 2026-10-11
trust unverified — no one has confirmed this page's content
sources 1 — jacob-levine-2026

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