CVPR 2026 was held in Denver, Colorado, bringing together thousands of researchers, engineers, and industry professionals from around the world. CVPR featured a broad range of technical sessions, keynote talks, tutorials, workshops, demonstrations, and industry exhibitions.
I had the opportunity to present our paper, HCL-FF: Hierarchical and Contrastive Learning for Forward-Forward Algorithm, in the poster session. The work introduces a novel framework that combines hierarchical representation learning with contrastive objectives to improve the Forward-Forward algorithm, an alternative learning paradigm that seeks to move beyond conventional backpropagation. Throughout the session, I had engaging discussions with researchers working on biologically inspired learning, efficient neural network training, and alternative optimization methods.
Beyond the technical presentations, CVPR offered numerous networking opportunities. The industry exhibition featured demonstrations from leading technology companies and AI startups, showcasing state-of-the-art developments in computer vision applications ranging from autonomous systems and robotics to generative media and multimodal AI. The conference environment encouraged interactions between academic researchers and industry practitioners, creating a vibrant atmosphere for exchanging ideas.
