Exploring the role of care in educational systems

Team-based problem framing with Bloombox tools during the gathering.

Over the past weeks, I found myself moving between two different and deeply connected spaces in the UK.

At the University of Cambridge, I gave a talk at the THRiVE research group on participatory and embodied approaches to learning. A few days later, I joined the Bloombox gathering, a small retreat bringing together educators, researchers, and theologians to explore care and the deeper purposes of education.

One space was structured around research, methods, and conceptual clarity. The other unfolded through dialogue, presence, and shared inquiry. Beneath these differences, both seemed to circle the same question:

How do we learn — and design systems — not only to solve problems, but to remain together in the face of difference, uncertainty, and conflict?

Continue reading

Integrating Behavior, Text, and Networks to Forecast Online Participation

As online platforms increasingly rely on voluntary contributions—from open science to collaborative innovation—the ability to anticipate user engagement becomes both a scientific and practical priority. Yet predicting who will stay active, who will disengage, and why, remains a complex challenge. Our recent paper, KEGNN: Knowledge-Enhanced Graph Neural Networks for User Engagement Prediction (Fan et al., International Conference on Multimedia Retrieval 2025), introduces a novel framework that addresses this gap by integrating behavioral, social, and semantic signals into a unified predictive model.

Continue reading

Teams That Thrive: AI-Driven Collaboration for Youth Participation

Team success isn’t just about outcomes—it’s also about how people feel, relate, and engage along the way. Understanding and improving this human dimension of participation is key to building teams that flourish. That’s the question we set out to answer in our latest study in the journal Computers and education: Artificial Intelligence, a collaboration between the Artificial Intelligence Research Institute in Spain and our team at the Learning Planet Institute in Paris. Together, we explored how artificial intelligence can help compose better teams in Challenge-Based Learning (CBL) environments, with a special focus on participants’ experiences—what we call participation quality.

Continue reading