Skip to content

MALA (Modular Artificial Learning Assistance)

Product or Programme · All products and programmes
Evidence · 1 claim (1 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

A GPT-4o-based modular chatbot architecture developed by Gabelmann et al. that routes exercise-solving requests among hint, explanation, feedback, and safety modules.

Components

  • MALA: a modular agentic AI chatbot architecture for exercise-solving assistance: MALA (Modular Artificial Learning Assistance) decomposes tutoring into a two-stage process of intent classification and task-specific execution. A central classifier routes each student message to one of four specialized modules: a hint module that handles requests where students ask for 'a starting point, the next step, or a nudge toward the solution', an explanation module constrained to short structured messages, a feedback module with a reasoning-then-evaluation-then-output workflow, and a fallback/safety module deflecting prompt-injection attempts. It was prototyped with GPT-4o for guiding students through problem-solving exercises and optionally generating new exercises. (Gabelmann et al. (2026))

Claims

Key Sources

  • Gabelmann, J., Jahn, F., Baum, K., van Rossum, S., Wuenscher, E., Gros, T. P., Wolf, V. (2026). Modularizing Educational LLM-Agency for Fostering Responsible Learning Assistance. https://arxiv.org/abs/2605.30187. doi:10.48550/arxiv.2605.30187
Page metadata
Field Value
type product
id mala-modular-artificial-learning-assistance
title MALA (Modular Artificial Learning Assistance)
description A GPT-4o-based modular chatbot architecture developed by Gabelmann et al.
status review
product_kind software
generated by process:settle-candidates on 2026-10-11
trust unverified — no one has confirmed this page's content

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