This six-volume set LNAI constitutes the refereed proceedings of the 27th International Conference on Artificial Intelligence in Education, AIED 2026, held in Seoul, South Korea, during June 27July 3, 2026.

The 143 full papers and 165 short papers presented in this book were carefully reviewed and selected from 1241 submissions.

• The conference program comprises seven thematic tracks:
  Track 1: Technical Aspects of AIED
  Track 2: Human Aspects of AIED
  Track 3: Societal Aspects of AIED
  This year's theme, "From tools to teammates: human-AI synergy for Augmented Learning" , highlights research on human and AI agency, collaborative intelligence, and human & AI co-evolving.

Inhalt

.- An AI-Assisted Co-Planning System for Early English Reading Practice.
.- Capture-Calibrate-Coach: A Graph-Based Framework for Knowledge Monitoring Estimation and Adaptive Feedback.
.- The Cost of Thinking: Increased Jailbreak Risk in Large Language Models in Education.
.- CLAIRE: a Controllable LLM Tutoring Framework for Reading Comprehension.
.- RoMathExam: A Longitudinal Dataset of Romanian Math Exams (18952025) with a Seven-Decade Core (19572025).
.- Offline Reinforcement Learning for Adaptive Feedback in Online Programming Education.
.- Scalable and Explainable Learner-Video Interaction Prediction using Multimodal Large Language Models.
.- Confidence Estimation in Automatic Short Answer Grading with LLMs.
.- Evaluating Interactivity: Toward Automated Assessment of AI-Generated Explorable Explanations.
.- MisEdu-RAG: A Misconception-Aware Dual-Hypergraph RAG for Novice Math Teachers.
.- SEPT Privacy-by-Design Framework for Early Teacher Intervention: Modeling Difficulty as a Deviation from a Student's Success Signature.
.- SLAI: AI Support for Multilingual Small Group Discussions in Science Classrooms.
.- Cold-Start Syntax Error Prediction in Programming Education: Comparing Sequential Knowledge Tracing and Large Language Models.
.- Methodologies for Improving the Quality of AI Tutoring in K-12 Education.
.- Small, Private Language Models as Teammates for Educational Assessment Design.
.- Using LLMs to Annotate Pedagogical Moves: You Know What I Mean?.
.- When to Stop? An Experimental Study on AI Teachable Agent Stopping Mechanisms and Their Learning Affordances in Mathematics.
.- Can We Trust AI's Self-Assessment? Evaluating and Improving LLM Confidence Calibration in Educational Dialogue Coding.
.- Learning Context Matters: Measuring and Diagnosing Personalization Gaps in AI Instructional Design.
.- Optimizing In-Context Demonstrations for LLM-based Automated Grading.
.- Modeling Completion Time in Mathematics Formative Assessments: Content-Based Prediction of Time Variation.
.- Modeling Student Learning with 3.8 Million Program Traces.
.- CLARA: An AI-Augmented Analytics Dashboard for Collaboration Literacy.
.- SimPath: Clinically Grounded AI Patients for Therapist Training.
.- MusicTutor: Facilitating Goal-Oriented Singing Practice via Multi-Agent Tutoring Framework.
.- The Missing Evaluation Axis: What 10,000 Student Submissions Reveal About AI Tutor Effectiveness.
.- Learning in Blocks: A Multi Agent Debate Assisted Personalized Adaptive Learning Framework for Language Learning.
.- Can Multimodal LLMs 'See' Science Instruction? Benchmarking Pedagogical Reasoning in K12 Classroom Videos.
.- RL Agents Reveal What's Hard: Bootstrapping Difficulty-Ordered Curricula for Human Learners.
.- Towards Real-Time Personalized Feedback in Open-Ended Learning Environments.
.- 4OPS: Exact Enumeration and Difficulty Modeling for Integer Arithmetic Puzzles.
.- Using Poly-Encoders for Computationally Efficient Automated Creativity Assessment.
.- Evaluating Vision-Language and Large Language Models for Automated Student Assessment in Indonesian Classrooms.
.- From writing traces to personalised support: Guiding LLMs with stylometric fingerprints.
.- DebugTA: An LLM-Based Agent for Simplifying Debugging and Teaching in Programming Education.
.- FoundationalASSIST: Dataset for Foundational Knowledge Tracing & Pedagogical Grounding of Large Language Models.
.- Leveraging LLMs for Dynamic Engagement Pattern Recognition in Collaborative Learning.
.- THiNK: Can Large Language Models Think-aloud?.
.- Student Development Agent: Risk-free Simulation for Evaluating AIED Innovations.
.- Design, Integration, and Evaluation of LLM-Enhancement Techniques for LMSs: RAG, LoRA Fine-Tuning, and Structured Generation.
.- Multi-Dimensional Evaluation of LLMs for Grammatical Error Correction.
.- From BOPPPS Stages to Cognitive-Adaptive Prompts: Controlling Instructional Drift in LLM-Based Tutoring Dialogues.
.- Measuring What Matters---or What's Convenient?: Robustness of LLM-Based Scoring Systems to Construct-Irrelevant Factors.
.- The Limits of Text-Only LLMs for Think-Aloud Transcript Correction: A Multimodal Comparison in the Context of a Puzzle Video Game.
.- Using LLMs for Knowledge Component-level Correctness Labeling in Open-ended Coding Problems.
.- Translating XAI into Actionable Feedback Using LLMs to Prevent Student Dropout.
.- AI Roles as Multiple-Source Texts: Designing a Detective-Inquiry Writing Activity for Elementary Students.
.- LLM-Driven Text Simplification and its Effects on Extensive Reading in EFL Learners.
.- Predicting the Finish Before the Draft Ends: Continuous Forecasting of Writing Performance from Process Traces.
.- MPKT: A Knowledge Tracing Model for Music Performance Learning.
.- Automated Grading of Handwritten Mathematics Using Vision-Capable LLMs.
.- How Prompting and Context Granularity Shape LLM-Based Assessment of Academic Writing: A Case Study in Serbian.