Prompt-Engineering Framework for Real-Time Personalization in LLM-Based AI Teaching Assistants

Forum topic · 小凯 · 2026-09-06

Summary

A study by Saptarshi Basu, Sandeep Kakar, and Ashok Goel (arXiv:2509.00005) proposes a prompt-engineering framework to add scalable, flexible, real-time micro-level personalization to general-purpose LLM/RAG-based AI teaching assistants such as Jill Watson, without model retraining. The framework adapts responses along six learner-specific dimensions: self-assessment, abstraction preference, verbosity preference, perceptual orientation, information processing style, and level of understanding, yielding 96 distinct learner profiles. Each student query is also analyzed using Bloom's Taxonomy to estimate cognitive complexity at the interaction level. Learner attributes and cognitive assessments are encoded into structured prompts that condition the LLM's output. The framework was evaluated using NLP metrics and a human study with five participants. Results show perceived differences in response style and structure under personalized conditions, and statistical analysis identified learner attributes associated with measurable response variation. The authors present this as preliminary evidence that prompt-based personalization can support adaptive behavior in LLM-driven educational agents.

Overview

Field: AI/ML Authors: Saptarshi Basu, Sandeep Kakar, Ashok Goel Published: 2026-09-06 arXiv: 2509.00005

Abstract (translation)

AI teaching assistants powered by large language models (LLMs) offer scalable educational support but often provide limited personalization. This study presents a prompt-engineering-based framework for personalizing general-purpose LLM/RAG-based AI teaching assistants such as Jill Watson across academic disciplines and courses. The framework adapts responses using six learner-specific dimensions:

Self-assessment
Abstraction preference
Verbosity preference
Perceptual orientation
Information processing style
Level of understanding

Together these dimensions yield 96 distinct learner profiles. Student queries are additionally analyzed using Bloom's Taxonomy to estimate cognitive complexity at the interaction level. Learner attributes and cognitive assessments are encoded in structured prompts, conditioning the LLM without retraining the model.

The framework was experimentally evaluated using NLP metrics and a human study with five participants. Results show perceived differences in response style and structure under personalized conditions, and statistical analysis identified learner attributes associated with measurable response variation. These findings provide preliminary evidence that prompt-based personalization can support adaptive behavior in LLM-driven educational agents.

Key takeaways

Micro-level personalization is achieved purely through structured prompting — no fine-tuning or retraining required.
Six learner dimensions combine into 96 learner profiles, enabling hybrid personalization across disciplines and courses.
Bloom's Taxonomy is used per-interaction to gauge the cognitive complexity of each student query.
Evaluation combines automated NLP metrics with a small human study, showing measurable stylistic differences and statistically linked learner attributes.

--- *Auto-collected on 2026-09-06*