AdaPT

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

Description

AdaPT is a Vue.js and TypeScript NestJS web application developed by Yanjie Zhang and colleagues that structures existing lesson plans, analyzes student learning profiles, and supports teacher-in-the-loop adaptation for differentiated K–12 instruction.

Components

• Multi-agent backend with LessonPlan Agent and LearningProfile Analysis Agent providing traceable transformations: AdaPT's backend is a multi-agent system in which each agent encapsulates distinct reasoning responsibilities. The LessonPlan Agent parses uploaded plans into schema-conformant components (using RAG over a tree-structured knowledge graph of national curriculum standards) and transforms them by recalibrating objectives and iteratively adjusting tasks. The LearningProfile Analysis Agent generates structured profiles from teacher natural-language descriptions, optionally retrieving knowledge readiness from the school's LMS via tool-calling. The authors state "The backend of AdaPT is designed as a multi-agent system (Figure 5), where each agent encapsulates distinct reasoning responsibilities aligned with the design requirements." Outputs are schema-validated JSON with references and localized reasons for traceability. (Yanjie Zhang et al. (2026))

Claims

• A formative study with six teachers and two experts identified five core challenges in lesson plan adaptation, including lack of systematic student profile data and fragmented refinement workflows [+M]

Related Products and Programmes

-

Key Sources

• Yanjie Zhang, Jiajun Zhu, Minyu Wu, Huamin Qu, and Sicheng Song. (2026). AdaPT: Adaptive Lesson Plan Transformer for Cross-Regional and Differentiated Instruction. https://arxiv.org/abs/2606.17633

Page metadata

Field	Value
type	product
id	adapt
title	AdaPT
description	AdaPT is a Vue.js and TypeScript NestJS web application developed by Yanjie Zhang and colleagues that structures existing lesson plans, analyzes student learning profiles, and supports teacher-in-the-loop adaptation for differentiated K–12 instruction.
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.