Steering Technique
Steering techniques encompass methods for guiding complex systems towards desired outcomes, ranging from autonomous vehicle navigation to controlling language models and microrobots. Current research emphasizes developing robust and efficient steering strategies using diverse approaches, including model-based reinforcement learning, neural network architectures (like CNNs and LSTMs), and optimization algorithms (e.g., gradient-based methods and iterative LQR). These advancements have significant implications for various fields, improving the safety and performance of autonomous systems, enhancing human-computer interaction in creative tasks, and enabling more precise control in medical procedures.
Papers
SteeredMarigold: Steering Diffusion Towards Depth Completion of Largely Incomplete Depth Maps
Jakub Gregorek, Lazaros Nalpantidis
Human Insights Driven Latent Space for Different Driving Perspectives: A Unified Encoder for Efficient Multi-Task Inference
Huy-Dung Nguyen, Anass Bairouk, Mirjana Maras, Wei Xiao, Tsun-Hsuan Wang, Patrick Chareyre, Ramin Hasani, Marc Blanchon, Daniela Rus