
Dear colleagues, We are thrilled to announce the official start of the first BabyBench Competition featured at the IEEE International Conference on Development and Learning (ICDL) 2025! The topic of this first edition will be how infants discover their own bodies. Can you help the multimodal infant model MIMo learn two typical infant behaviors: self touch and hand regard? What is BabyBench? BabyBench is a Multimodal Benchmark of Infant Behaviors for Developmental Artificial Intelligence. The goal of the competition is to devise learning algorithms that allow MIMo to obtain important milestones of cognitive development. We provide the embodiment, the simulation environments, and the evaluation metrics; all you need is to implement your ideas. Are you up for the challenge? Important dates • Submission deadline: September 1, 2025 • Winner announcement at ICDL: September 19, 2025 Why Participate? • Low Entry Barrier: Get started quickly using our pre-defined, user-friendly simulation environment. Whether you're a seasoned researcher or a curious newcomer, the BabyBench platform is designed to let you focus on modeling, not engineering. • The winner will receive a €150 prize sponsored by the IEEE RAS TC Cognitive Robotics. • All finalists will receive a certificate at the IEEE ICDL. • Contribute to the Developmental AI community! For more information visit https://babybench.github.io/2025 Main contact point: Francisco Lopez (fcomlop@gmail.com) Feel free to forward this to anyone who might be interested! Warm regards, The BabyBench 2025 Organizing Team Francisco M. López (Frankfurt Institute for Advanced Studies) Valentin Marcel (Czech Technical University in Prague) Xavier Hinaut (INRIA Bordeaux) Jochen Triesch (Frankfurt Institute for Advanced Studies) Matej Hoffmann (Czech Technical University in Prague)