Join us with our workshop: Vision Language Models: Challenges of RealWorld Deploymen
The WISE Lab is thrilled to be hosting our inaugural event, the 1st Workshop on VLM4RWD: Vision Language Models: Challenges of Real-World Deployment, as part of the prestigious NeurIPS 2025 conference. We are especially excited to welcome the global research community to this full-day workshop in the vibrant setting of Mexico City, Mexico, on November 30th, 2025.
Vision Language Models (VLMs) have shown immense potential for applications like robotics and autonomous systems. However, their transition from experimental systems to reliable technologies is hindered by significant challenges in efficiency, scalability, and robustness—issues critical for deployment but often overlooked in traditional research.
This workshop, organized by Dr. Mozhgan Nasr Azadani, Yimu Wang, Prof. Krzysztof Czarnecki, Dr. Negar Arabzadeh (UC Berkeley), Dr. Richard He Bai (Apple), Dr. Lukas Schmid (MIT), will bridge this gap. It aims to establish concrete pathways toward building VLMs that can operate reliably under practical constraints. The event will bring together researchers to discuss cutting-edge research, system-level optimizations, and evaluation methodologies, emphasizing efficiency, robustness, and reliability as core design principles. It will feature distinguished speakers, poster sessions, and stimulating discussions that push the boundaries of VLM deployment.
We are genuinely looking forward to meeting people and engaging with the community in Mexico City to share insights and foster collaborations that will define the future of reliable Vision Language Systems.