Enhancing Diversity of LLM-Generated Educational Tasks

2025/12/29 by Manh Hung Nguyen, Tschiatschek, Sebastian, Nguyen, Manh Hung +1
Computer Science · Psychology · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Games #FOS: Computer and information sciences #Innovative Teaching and Learning Methods #Intelligent Tutoring Systems and Adaptive Learning

paper · doi:10.48550/arxiv.2512.23601

openalex publication_date 2025/12/29 · openalex created_date 2025/12/31 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) have shown the potential for generating educational content at scale, assisting educators in creating practice tasks or synthesizing data for training educational models. However, LLMs suffer from the ``Artificial Hivemind'' effect, where they produce homogeneous content. This homogeneity limits the diversity of LLM-generated tasks, a crucial factor in these educational settings. In this paper, we investigate how to increase the diversity of generated tasks while keeping their utility high. Inspired by the divergent--convergent thinking stages in creativity literature, we propose a prompting framework with two reasoning stages: (1) exploring the creative space, and (2) satisfying the input requirements. We evaluate CreativeDC, a method instantiated from this framework in the domain of Python programming, using both automated metrics and expert evaluation. Results show that CreativeDC produces significantly more distinct high-utility tasks (about 1.6×) than baselines. Our work offers an effective approach for generating and evaluating more diverse tasks at scale.

Citations

• Multi-Agent Collaborative Framework For Math Problem Generation
• Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)
• Prompt Optimization Across Multiple Agents for Representing Diverse Human Populations
• Partnering with AI: A Pedagogical Feedback System for LLM Integration into Programming Education
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
• Creative Preference Optimization
• From Recall to Reasoning: Automated Question Generation for Deeper Math Learning through Large Language Models
• Qwen3 Technical Report
• Measuring LLM Novelty As The Frontier Of Original And High-Quality Output
• Generating Planning Feedback for Open-Ended Programming Exercises with LLMs
• Synthesizing High-Quality Programming Tasks with LLM-based Expert and Student Agents
• NoveltyBench: Evaluating Language Models for Humanlike Diversity
• Training LLM-based Tutors to Improve Student Learning Outcomes in Dialogues
• A Novel Approach to Scalable and Automatic Topic-Controlled Question Generation in Education
• AI as Humanity's Salieri: Quantifying Linguistic Creativity of Language Models via Systematic Attribution of Machine Text against Web Text
• Scaling Synthetic Data Creation with 1,000,000,000 Personas
• Evaluating Contextually Personalized Programming Exercises Created with Generative AI
• LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play
• Math Multiple Choice Question Generation via Human-Large Language Model Collaboration
• Generative AI for Education (GAIED): Advances, Opportunities, and Challenges
• How Teachers Can Use Large Language Models and Bloom's Taxonomy to Create Educational Quizzes
• Scalable Educational Question Generation with Pre-trained Language Models
• The Vendi Score: A Diversity Evaluation Metric for Machine Learning
• BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence
• Math Word Problem Generation with Mathematical Consistency and Problem Context Constraints

Related

• Creative Problem Solving in Large Language and Vision Models -- What Would it Take?
• CreativityPrism: A Cross-Domain Evaluation Framework for Large Language Model Creativity
• Human Creativity in the Age of LLMs: Randomized Experiments on Divergent and Convergent Thinking
• Universe of Thoughts: Enabling Creative Reasoning with Large Language Models
• Large Language Models for Scientific Idea Generation: A Creativity-Centered Survey