CODE-GEN achieves human-validated success rates of 79.9% to 98.6% across seven pedagogical evaluation dimensions¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Five of seven dimensions exceeded 91% human-validated success with failure rates of 3.1% or below, while success rates ranged from 79.9% to 98.6% across dimensions. → Xiaojing Duan 2026
Evidence¶
Xiaojing Duan 2026¶
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 doi:10.48550/arxiv.2604.03926
q2 · i? · design · r2
Evaluation study in which six human subject-matter experts judged 288 AI-generated questions, yielding 2,016 SME-Validator rated pairs treated as ground truth in a binary classification framework. The article reports "human-validated success rates exceeding 91%" on five of seven dimensions with "minimal failure rates of 3.1% or below"; no effect size is printed.
"Five of seven dimensions achieved human-validated success rates exceeding 91%, accompanied by minimal failure rates of 3.1% or below."
Discussion¶
Related Claims¶
- Concept alignment achieves the highest success rate (98.6%) among the seven dimensions, attributed to the RAG-based design — a narrower finding that bears on this claim
- Distractor quality is the weakest automated dimension, with the lowest success rate (79.9%) and highest failure rate (15.6%) — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
code-gen-success-rates-seven-dimensions |
title |
CODE-GEN achieves human-validated success rates of 79.9% to 98.6% across seven pedagogical evaluation dimensions |
description |
CODE-GEN achieves human-validated success rates of 79.9% to 98.6% across seven pedagogical evaluation dimensions |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — xiaojing-duan-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.