Aladin Djuhera, Haris Gacanin, and Holger Boche received the Best Paper Award for their paper, “MambaCSP: Hybrid-Attention State Space Models for Hardware-Efficient Channel State Prediction” at the workshop “AI and ML for Next-Generation Wireless Communications and Networking” held as part of the International Conference on Machine Learning (ICML) in Seoul, South Korea, in July 2026.
The International Conference on Machine Learning (ICML) holds the highest A* and Tier 1 rankings in computer science. Alongside NeurIPS and ICLR, it is widely regarded as one of the “Big Three” conferences in machine learning. It is an extremely prestigious and sought-after forum for AI research.
With the emergence of NextG wireless technologies such as 6G, artificial intelligence is on the verge of fundamentally redefining network design and enabling next-generation physical AI applications—yet a critical knowledge gap still exists between the research communities in the fields of machine learning and wireless communications. To bridge this gap, the second AI4NextG workshop brought together experts from academia and industry to develop domain-optimized, AI-native solutions across the entire wireless protocol stack. Building on the success of the inaugural event, this workshop aimed to advance deployable innovations that enable the ultra-fast, intelligent connectivity required for robotics, autonomous systems, and the future of physical AI.
