11 papers
From Substitution to Scaffolding: Breaking the Self-Reinforcing Harm Cycle of AI in Education (and Beyond)
Lucile Favero, Juan Antonio Pérez-Ortiz, Tanja Käser +1
Artificial intelligence is being adopted in educational settings faster than its consequences are understood. We argue that the central risk is misalignment: AI that eliminates hum…
PIVOT: Preference-based Intervention Vectors for Pedagogical Tutor Steering
Fares Fawzi, Jiaxu Zhao, Tanya Nazaretsky +1
LLMs are increasingly used for conversational tutoring, but effective tutoring requires more than correct answers. Tutors must choose when to scaffold reasoning, hint, give feedbac…
Evaluating Answer Leakage Robustness of LLM Tutors against Adversarial Student Attacks
Jin Zhao, Marta KneževiÄ, Tanja Käser
Large Language Models (LLMs) are increasingly used in education, yet their default helpfulness often conflicts with pedagogical principles. Prior work evaluates pedagogical quality…
Scalable and Explainable Learner-Video Interaction Prediction using Multimodal Large Language Models
Dominik Glandorf, Fares Fawzi, Tanja Käser
Learners' use of video controls in educational videos provides implicit signals of cognitive processing and instructional design quality, yet the lack of scalable and explainable p…
REFINE: Real-world Exploration of Interactive Feedback and Student Behaviour
Fares Fawzi, Seyed Parsa Neshaei, Marta Knezevic +2
Formative feedback is central to effective learning, yet providing timely, individualised feedback at scale remains a persistent challenge. While recent work has explored the use o…
Moving Beyond Review: Applying Language Models to Planning and Translation in Reflection
Seyed Parsa Neshaei, Richard Lee Davis, Tanja Käser
Reflective writing is known to support the development of students' metacognitive skills, yet learners often struggle to engage in deep reflection, limiting learning gains. Althoug…