ProPACT: A Proactive AI-Driven Adaptive Collaborative Tutor for Pair Programming
Anahita Golrang, Kshitij Sharma, olga viberg
TLDR
ProPACT is a proactive AI tutor that uses multimodal data and XGBoost to predict and improve pair programming collaboration in real-time.
Key contributions
- Introduces ProPACT, a proactive AI tutor that teaches collaboration in pair programming.
- Employs a multimodal dyadic model and XGBoost to predict collaboration issues 30 seconds in advance.
- Delivers minimally intrusive, adaptive scaffolds and fades support during productive collaboration.
- Study shows ProPACT improves debugging success, task efficiency, and feedback uptake in dyads.
Why it matters
This paper addresses a gap in adaptive learning by focusing on collaborative dynamics rather than individuals. By proactively identifying and addressing suboptimal collaboration states, ProPACT offers a novel approach to real-time learning support. Its success demonstrates the potential for AI to enhance complex social learning activities.
Original Abstract
Effective pair programming depends on coordination of attention, cognitive effort, and joint regulation over time, yet most adaptive learning systems remain individual-centric and reactive. This paper introduces ProPACT, a proactive AI-driven adaptive collaborative tutor that treats collaboration itself as the object of instruction. ProPACT constructs a multimodal dyadic learner model based on Joint Visual Attention (JVA), Joint Mental Effort (JME), and individual mental effort, and employs an XGBoost-based forecasting model to predict emerging suboptimal collaboration states up to 30 seconds in advance. These predictions drive a hierarchical adaptive policy that delivers minimally intrusive scaffolds while fading support during productive collaboration. A within-subject study with 26 pair-programming dyads shows that proactive feedback significantly improves debugging success, task efficiency, feedback uptake, and post-intervention gains in JVA and JME, demonstrating the potential of forecast-driven dyadic adaptivity for real-time collaborative learning regulation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.