skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


This content will become publicly available on July 1, 2026

Title: Multi-Agentic LLMs for Personalizing STEM Texts
Multi-agent large language models promise flexible, modular architectures for delivering personalized educational content. Drawing on a pilot randomized controlled trial with middle school students (n = 23), we introduce a two-agent GPT-4 framework in which a Profiler agent infers learner-specific preferences and a Rewrite agent dynamically adapts science passages via an explicit message-passing protocol. We implement structured system and user prompts as inter-agent communication schemas to enable real-time content adaptation. The results of an ordinal logistic regression analysis hinted that students may be more likely to prefer texts aligned with their profile, demonstrating the feasibility of multi-agent system-driven personalization and highlighting the need for additional work to build upon this pilot study. Beyond empirical validation, we present a modular multi-agent architecture detailing agent roles, communication interfaces, and scalability considerations. We discuss design best practices, ethical safeguards, and pathways for extending this framework to collaborative agent networks—such as feedback-analysis agents—in K-12 settings. These results advance both our theoretical and applied understanding of multi-agent LLM systems for personalized learning.  more » « less
Award ID(s):
2120888
PAR ID:
10639213
Author(s) / Creator(s):
; ;
Publisher / Repository:
MDPI
Date Published:
Journal Name:
Applied Sciences
Volume:
15
Issue:
13
ISSN:
2076-3417
Page Range / eLocation ID:
7579
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; (, Proceedings of the 15th International Conference on Educational Data Mining)
    Mitrovic, A.; & Bosch, N. (Ed.)
    Working collaboratively in groups can positively impact performance and student engagement. Intelligent social agents can provide a source of personalized support for students, and their benefits likely extend to collaborative settings, but it is difficult to determine how these agents should interact with students. Reinforcement learning (RL) offers an opportunity for adapting the interactions between the social agent and the students to better support collaboration and learning. However, using RL in education with social agents typically involves training using real students. In this work, we train an RL agent in a high-quality simulated environment to learn how to improve students’ collaboration. Data was collected during a pilot study with dyads of students who worked together to tutor an intelligent teachable robot. We explore the process of building an environment from the data, training a policy, and the impact of the policy on different students, compared to various baselines. 
    more » « less
  2. (, Proceedings of the ... AAAI Conference on Artificial Intelligence)
    Automated feedback has the potential to provide sig- nificant assistance to student game creators. Here, we present a system for generating automated, critique- like feedback for students creating games in the Study- Crafter platform. We implemented a system that builds a personalized feedback report for students based on a templated format. This critique uses automated analysis of structural and interactive aspects of the game narra- tive and recommends alternate games for students to ex- amine as inspiration. To test our system, we conducted a pilot study with 10 student groups developing narrative- based games. A key understanding from the study is that determining the appropriate depth of assessment and critique without overwhelming the student is important. 
    more » « less
  3. ; ; (, American Society of Mechanical Engineers)
    In multi-agent Bayesian optimization for Design Space Exploration (DSE), identifying a communication network among agents to share useful design information for enhanced cooperation and performance, considering the trade-off between connectivity and cost, poses significant challenges. To address this challenge, we develop a distributed multi-agent Bayesian optimization (DMABO) framework and study how communication network structures/connectivity and the resulting cost would impact the performance of a team of agents when finding the global optimum. Specifically, we utilize Lloyd’s algorithm to partition the design space to assign distinct regions to individual agents for exploration in the distributed multi-agent system (MAS). Based on this partitioning, we generate communication networks among agents using two models: 1) a range-limited model of communication constrained by neighborhood information; and 2) a range-free model without neighborhood constraints. We introduce network density as a metric to quantify communication costs. Then, we generate communication networks by gradually increasing the network density to assess the impact of communication costs on the performance of MAS in DSE. The experimental results show that the communication network based on the range-limited model can significantly improve performance without incurring high communication costs. This indicates that increasing the density of a communication network does not necessarily improve MAS performance in DSE. Furthermore, the results indicate that communication is only beneficial for team performance if it occurs between specific agents whose search regions are critically relevant to the location of the global optimum. The proposed DMABO framework and the insights obtained can help identify the best trade-off between communication structure and cost for MAS in unknown design space exploration. 
    more » « less
  4. ; ; ; ; ; ; (, Journal of Problem Based Learning in Higher Education)
    With the emergence of data science as an inherently multidisciplinary subject, there is increasing demand for graduates with well-rounded competence in computing, analytics, and communication skills. However, in conventional education systems, computing & quantitative, and communication skills are often taught in different disciplines. Data storytelling is constructing and presenting data stories to highlight the analytical insights to achieve the communication goals to a specific audience. Digital data storytelling leverages digital storytelling techniques and best practices in communication to deliver stories that can be shared in digital formats to a wide audience. In this paper, we describe and reflect on a semester-long project-based learning pilot using Digital Storytelling as a framework to allow students to explore topics themed around human flourishing and sustainability with the end goal of constructing data stories delivered in digital or video format (i.e., Digital Data Storytelling). The pilot work was conducted in an introductory data science course at a 4-year Minority Serving Institution in collaboration with students studying non-STEM disciplines at a partner community college. Our pilot demonstrates the potential benefit of this sustainability-aware Project-Based Learning design in raising students’ awareness of sustainability issues, increasing confidence in cross-disciplinary communication competency, and at the same time deepening their understanding of data science concepts. We further reflect on the significant role of an effective program model as well as challenges and opportunities for building transdisciplinary communication competency to prepare for a diverse data science workforce. 
    more » « less
  5. ; ; (, British Journal of Educational Technology)
    With increasing interest in computer‐assisted educa‐ tion, AI‐integrated systems become highly applicable with their ability to adapt based on user interactions. In this context, this paper focuses on understanding and analysing first‐year undergraduate student responses to an intelligent educational system that applies multi‐agent reinforcement learning as an AI tutor. With human–computer interaction at the centre, we discuss principles of interface design and educational gamification in the context of multiple years of student observations, student feedback surveys and focus group interviews. We show positive feedback from the design methodology we discuss as well as the overall process of providing automated tutoring in a gamified virtual environment. We also discuss students' thinking in the context of gamified educational systems, as well as unexpected issues that may arise when implementing such systems. Ultimately, our design iterations and analysis both offer new insights for practical implementation of computer‐assisted educational systems, focusing on how AI can augment, rather than replace, human intelligence in the classroom. Practitioner notesWhat is already known about this topicAI‐integrated systems show promise for personalizing learning and improving student education.Existing research has shown the value of personalized learner feedback.Engaged students learn more effectively.What this paper addsStudent opinions of and responses to an HCI‐based personalized educational system.New insights for practical implementation of AI‐integrated educational systems informed by years of student observations and system improvements.Qualitative insights into system design to improve human–computer interaction in educational systems.Implications for practice and/or policyActionable design principles for computer‐assisted tutoring systems derived from first‐hand student feedback and observations.Encourage new directions for human–computer interaction in educational systems. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email