Skip to content

Control timing as a design dimension for AI-assisted visual authoring: pre-, mid-, and post-generation control

Design · All designs
Evidence · 2 claims (2 for) · 1 study (1 causal), q3 · 0 of 1 report an effect size · 2 claims rest on one study

Description

The article introduces "control timing" as a key design dimension in AI-assisted visual authoring, operationalized through three interaction paradigms that differ in when teachers intervene relative to AI output. Pre-generation control means users "specify intent solely through natural language prompts before generation"; mid-generation control means users "inspect and confirm an explicit layout structure before the system completes generation"; post-generation control means users "directly modify AI-generated visuals after generation through object-level edits". The framework holds task content constant and varies only the timing of human control.

Design Implications

Context

Requirements

  • The paradigms require a multi-stage generation pipeline with distinct points (intent specification, intermediate planning, localized refinement) at which control can be exercised.

Constraints

  • The article instantiates and evaluates the three paradigms only for primary mathematics word problems in a single-session study.

Target Learners

  • primary mathematics teachers (Grades 1–6)

Learning Goals

  • creating pedagogically correct instructional visuals for arithmetic word problems

Claims

Examples

-

Key Sources

  • Zhengxu Li, Junling Wang, and April Yi Wang. (2026). When Should Teachers Control AI Generation for Mathematics Visuals? arXiv preprint. https://arxiv.org/abs/2605.10672
Page metadata
Field Value
type design
id control-timing-design-space-ai-visual-authoring
title Control timing as a design dimension for AI-assisted visual authoring: pre-, mid-, and post-generation control
description The article introduces "control timing" as a key design dimension in AI-assisted visual authoring, operationalized through three interaction paradigms that differ in when teachers intervene relative to AI output.
status review
generated by process:settle-candidates on 2026-10-12
trust unverified — no one has confirmed this page's content
sources 1 — zhengxu-li-2026

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