"AI just keeps guessing": Using ARC Puzzles to Help Children Identify Reasoning Errors in Generative AI

• URL: http://arxiv.org/abs/2505.16034v1
• Date: Wed, 21 May 2025 21:27:23 GMT
• Title: "AI just keeps guessing": Using ARC Puzzles to Help Children Identify Reasoning Errors in Generative AI
• Authors: Aayushi Dangol, Trushaa Ramanan, Runhua Zhao, Julie A. Kientz, Robert Wolfe, Jason Yip,
• Abstract summary: The integration of generative Artificial Intelligence (genAI) into everyday life raises questions about the competencies required to critically engage with these technologies.<n>Unlike visual errors in genAI, textual mistakes are often harder to detect and require specific domain knowledge.<n>We developed AI Puzzlers, an interactive system based on the Abstraction and Reasoning Corpus (ARC), to help children identify and analyze errors in genAI.
• Score: 4.80495766531247
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The integration of generative Artificial Intelligence (genAI) into everyday life raises questions about the competencies required to critically engage with these technologies. Unlike visual errors in genAI, textual mistakes are often harder to detect and require specific domain knowledge. Furthermore, AI's authoritative tone and structured responses can create an illusion of correctness, leading to overtrust, especially among children. To address this, we developed AI Puzzlers, an interactive system based on the Abstraction and Reasoning Corpus (ARC), to help children identify and analyze errors in genAI. Drawing on Mayer & Moreno's Cognitive Theory of Multimedia Learning, AI Puzzlers uses visual and verbal elements to reduce cognitive overload and support error detection. Based on two participatory design sessions with 21 children (ages 6 - 11), our findings provide both design insights and an empirical understanding of how children identify errors in genAI reasoning, develop strategies for navigating these errors, and evaluate AI outputs.

Related papers

• Children's Mental Models of AI Reasoning: Implications for AI Literacy Education [8.996593596034506]
  We identify three models of AI reasoning: Deductive, Inductive, and Inherent.<n>Our findings reveal that younger children (grades 3-5) often attribute AI's reasoning to inherent intelligence, while older children (grades 6-8) recognize AI as a pattern recognizer.<n>We highlight three tensions that surfaced in children's understanding of AI reasoning and conclude with implications for scaffolding AI curricula and designing explainable AI tools.
  arXiv  Detail & Related papers  (2025-05-21T21:20:12Z)
• The Imitation Game for Educational AI [23.71250100390303]
  We present a novel evaluation framework based on a two-phase Turing-like test.<n>In Phase 1, students provide open-ended responses to questions, revealing natural misconceptions.<n>In Phase 2, both AI and human experts, conditioned on each student's specific mistakes, generate distractors for new related questions.
  arXiv  Detail & Related papers  (2025-02-21T01:14:55Z)
• AI in Education: Rationale, Principles, and Instructional Implications [0.0]
  Generative AI, like ChatGPT, can create human-like content, prompting questions about its educational role.<n>The study emphasizes deliberate strategies to ensure AI complements, not replaces, genuine cognitive effort.
  arXiv  Detail & Related papers  (2024-12-02T14:08:07Z)
• Do great minds think alike? Investigating Human-AI Complementarity in Question Answering with CAIMIRA [43.116608441891096]
  Humans outperform AI systems in knowledge-grounded abductive and conceptual reasoning. State-of-the-art LLMs like GPT-4 and LLaMA show superior performance on targeted information retrieval.
  arXiv  Detail & Related papers  (2024-10-09T03:53:26Z)
• Combining AI Control Systems and Human Decision Support via Robustness and Criticality [53.10194953873209]
  We extend a methodology for adversarial explanations (AE) to state-of-the-art reinforcement learning frameworks. We show that the learned AI control system demonstrates robustness against adversarial tampering. In a training / learning framework, this technology can improve both the AI's decisions and explanations through human interaction.
  arXiv  Detail & Related papers  (2024-07-03T15:38:57Z)
• Evaluating Large Vision-and-Language Models on Children's Mathematical Olympiads [74.54183505245553]
  A systematic analysis of AI capabilities for joint vision and text reasoning is missing in the current scientific literature.<n>We evaluate state-of-the-art LVLMs on their mathematical and algorithmic reasoning abilities using visuo-linguistic problems from children's Olympiads.<n>Our results show that modern LVLMs do demonstrate increasingly powerful reasoning skills in solving problems for higher grades, but lack the foundations to correctly answer problems designed for younger children.
  arXiv  Detail & Related papers  (2024-06-22T05:04:39Z)
• OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI [73.75520820608232]
  We introduce OlympicArena, which includes 11,163 bilingual problems across both text-only and interleaved text-image modalities.<n>These challenges encompass a wide range of disciplines spanning seven fields and 62 international Olympic competitions, rigorously examined for data leakage.<n>Our evaluations reveal that even advanced models like GPT-4o only achieve a 39.97% overall accuracy, illustrating current AI limitations in complex reasoning and multimodal integration.
  arXiv  Detail & Related papers  (2024-06-18T16:20:53Z)
• Towards Reconciling Usability and Usefulness of Explainable AI Methodologies [2.715884199292287]
  Black-box AI systems can lead to liability and accountability issues when they produce an incorrect decision. Explainable AI (XAI) seeks to bridge the knowledge gap, between developers and end-users.
  arXiv  Detail & Related papers  (2023-01-13T01:08:49Z)
• Seamful XAI: Operationalizing Seamful Design in Explainable AI [59.89011292395202]
  Mistakes in AI systems are inevitable, arising from both technical limitations and sociotechnical gaps. We propose that seamful design can foster AI explainability by revealing sociotechnical and infrastructural mismatches. We explore this process with 43 AI practitioners and real end-users.
  arXiv  Detail & Related papers  (2022-11-12T21:54:05Z)
• A User-Centred Framework for Explainable Artificial Intelligence in Human-Robot Interaction [70.11080854486953]
  We propose a user-centred framework for XAI that focuses on its social-interactive aspect. The framework aims to provide a structure for interactive XAI solutions thought for non-expert users.
  arXiv  Detail & Related papers  (2021-09-27T09:56:23Z)
• The Who in XAI: How AI Background Shapes Perceptions of AI Explanations [61.49776160925216]
  We conduct a mixed-methods study of how two different groups--people with and without AI background--perceive different types of AI explanations. We find that (1) both groups showed unwarranted faith in numbers for different reasons and (2) each group found value in different explanations beyond their intended design.
  arXiv  Detail & Related papers  (2021-07-28T17:32:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.