AI evidence extraction

A Theory of Adaptive Scaffolding for LLM-Based Pedagogical Agents

Clayton Cohn, Surya Rayala, Namrata Srivastava, Joyce Horn Fonteles, S. K. Jain, X. Luo et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

5/10

Relevance

0/4

Quality (LMQS)

Evidence

Citations

24.52

FWCI

Top 10%

Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.1609/aaai.v40i3.37154

Methodology & findings

Study design

Design science / artifact instantiation with qualitative evaluation.

Sample

not–determinable, 1 group

Primary method

not_provided

Main result

The study found that "Inquizzitor delivers high-quality assessment and interaction aligned with core learning theories, offering effective guidance that students value." This demonstrates the potential for theory-driven LLM integration in education and the ability of these systems to provide adaptive and principled instruction.

Reports effect sizes and confidence intervals.

Research paradigm

pragmatist/design-science

Author conclusions

The authors conclude that "This research demonstrates the potential for theory-driven LLM integration in education, highlighting the ability of these systems to provide adaptive and principled instruction."

Risk of bias

No explicit statement of blinding procedures; No mention of pre-registration or protocol transparency; Evaluation methodology not detailed in abstract - potential observer bias or selective reporting; No discussion of potential bias in LLM training data affecting assessment fairness

Open questions raised

The paper identifies a gap between current LLM systems used in classrooms and earlier intelligent tutoring systems, noting that "current LLM systems used in classrooms often lack the solid theoretical foundations found in earlier intelligent tutoring systems."
The paper identifies a gap between current LLM systems in classrooms and earlier intelligent tutoring systems: "current LLM systems used in classrooms often lack the solid theoretical foundations found in earlier intelligent tutoring systems." The authors address this by proposing a theory-grounded framework for adaptive scaffolding.
not_explicitly_provided

Data: not_statedCode: not_statedExtracted from: pdfAgreement 73%

Explore related topics

AI Literacy Training Doctoral Training Case Study