Skip to content

CODE-GEN

Product or Programme · All products and programmes
Evidence · 2 claims (2 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description

CODE-GEN is a retrieval-augmented, dual-agent educational AI system developed by Xiaojing Duan et al. that generates course-grounded coding-comprehension questions and uses an independent Validator agent and execution tools to assess their pedagogical quality.

Components

  • CODE-GEN dual-agent architecture separating generation from validation: CODE-GEN is a human-in-the-loop agentic AI system integrating retrieval-augmented generation with a dual-agent architecture: a Generator agent produces multiple-choice coding comprehension questions grounded in retrieved course context, and a Validator agent independently assesses each question across seven pedagogical dimensions. Both agents are augmented with an Arithmetic Expression Evaluator and a Sandboxed Python Runner. The article states the separation "allows the system to decouple the creative task of question generation from the analytical task of quality evaluation" and treats automated validation as an empirical object of study rather than an assumed capability. (Xiaojing Duan et al. (2026))

Claims

-

Key Sources

  • Xiaojing Duan, Frederick Nwanganga, and Chaoli Wang. (2026). CODE-GEN: A Human-in-the-Loop RAG-Based Agentic AI System for Multiple-Choice Question Generation. https://arxiv.org/abs/2604.03926
Page metadata
Field Value
type product
id code-gen
title CODE-GEN
description CODE-GEN is a retrieval-augmented, dual-agent educational AI system developed by Xiaojing Duan et al.
status review
product_kind software
generated by process:settle-candidates on 2026-10-10
trust unverified — no one has confirmed this page's content

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