Call for Papers

XAI-Ed @ LAK27

Closing the Interpretation-Action Loop: Actionable Explainable AI to Enhance Learning and Teaching

Explanation quality in education is still assessed mostly through model-side proxies, such as fidelity and agreement between explainers. Turning insight into action remains one of the major challenges of the analytics cycle. This fourth edition of XAI-Ed focuses on that gap through two questions:

1. How do explanations influence the learning process and learning outcomes?
2. How do explanations support differently positioned stakeholders (students, teachers, parents and caregivers, and educational leaders) in taking action?

Equity, fairness, and transparency run through both questions, since explanations determine who is able to scrutinise, contest, or act upon automated inferences.

Topics of interest

We invite research contributions, including system demonstrations, on topics including:

• Effects of explanations on learning processes, self-regulated learning, and learning outcomes.
• Theoretically grounded explanation, linking XAI and XLA to learning theory and learning design.
• Actionability of explanations and the path from explanation to educational intervention.
• Multi-step and co-constructed explanations with stakeholders (social XAI and XLA).
• Stakeholder-differentiated explanation design and communication of inferences.
• Explainability of LLM-based systems, including faithfulness and calibrated reliance.
• Interpretable models for learner modelling, knowledge tracing, and success prediction.
• Fairness auditing, explanatory equity, and contestability of automated inferences.
• Evaluation frameworks for XLA, and governance and institutional adoption of explainability.

Review process

All submissions undergo double-blind review by at least two programme committee members. The programme committee consists of the organisers and external experienced reviewers.

Submissions are invited in English, and the workshop is conducted in English. No prior knowledge is required beyond an interest in explainability in learning analytics.

Important dates

Submission format and the submission link will be announced here.