A3 Lab is a research group at Aarhus University advancing adaptive and agentic AI for reliable deployment under non-stationarity. We develop multimodal foundation models and test-time learning methods that can quantify uncertainty, and adapt safely on the fly—minimizing costly re-training while maintaining robustness, auditability, and governance in real-world, safety-critical settings.
We study multimodal foundation models (vision-language and beyond) for robust perception and decision making under distribution shift. Our work includes prompting and instruction following, interactive and agentic systems, and explainability and faithful reasoning so that outputs are not only accurate but also verifiable. We extend these models to physical AI through vision-language-action (VLA) models and world models for embodied agents.
We develop novel self-supervised learning algorithms to build stronger representations and better vision foundation models that transfer across tasks and domains. We also explore synthetic data and generative models to improve robustness, cover rare conditions, and reduce dependence on labelled data.
We design lightweight methods that update models at inference time using unlabelled or weak supervision. Key topics include domain generalisation and test-time adaptation, parameter-efficient continual updates, and unknown-class detection and novelty recognition to handle evolving and recurring domain shifts.
We investigate multi-agent coordination and planning for complex AI pipelines, including human–AI teaming and decision support. To enable practical deployment, we work on efficient model design (TinyML, distillation, efficient transformers) and rigorous evaluation protocols for reliability under drift.
🎉 [October 2026] Dr. Bozorgtabar has been awarded DKK 7.2 million by the Independent Research Fund Denmark (DFF) for the STORM project on soft tactile memory for safe physical AI, with co-PI Prof. Ahmad Rafsanjani (SDU).
🎉 [May 2026] Dr. Bozorgtabar has been awarded an AIAS Associate Fellowship at Aarhus University.
🎉 [May 2026] One paper has been accepted to ICML 2026.
🎉 [May 2026] Three papers have been accepted to MICCAI 2026.
