5 papers indexed
Sixu Li, Thomas Jacob Maranzatto, Jan Peszek, Trevor Teolis, Semih Akkoc, Konstantin Riedl, et al.
We study the inference-time behavior of deep linear encoder-only transformers through the lens of interacting particle systems. In this perspective, tokens are modeled as particles that interact dynamically through successive linear self-attention layers. We show that in embeddin…
Sixu Li, Nitesh Kumar, Reyshwanth Ganeshan, Sivakumar Rathinam, Swaroop Darbha, Yang Zhou
As autonomous vehicles (AVs) operate in increasingly dynamic traffic conditions, lateral control must be performed while longitudinal speed and acceleration vary. Yet many existing lateral controllers rely on constant-speed or operating-point-based assumptions, which can degrade…
Kexin Tian, Sixu Li, Keshu Wu, Yang Zhou, Zhengzhong Tu
Visual Object Tracking (VOT) and Moving Object Segmentation (MOS) are two fundamental tasks in computer vision that involve both spatial and temporal object dynamics. Existing methods rely predominantly on visual cues and thus often falter in real-world scenarios where object mot…
Sixu Li, Swaroop Darbha, Yang Zhou
Connected and automated vehicle (CAV) platooning promises gains in energy efficiency and traffic throughput and, most critically, in safety. These safety benefits hinge on string stability, which determines how disturbances propagate along a platoon. While longitudinal string sta…
Nitesh Kumar, Reyshwanth Ganeshan, Sixu Li, Sivakumar Rathinam, Swaroop Darbha
Cooperative localization using range-based measurements is critical for multi-robot systems operating in GPS-denied and unstructured environments. However, traditional cooperative approaches require sharing explicit spatial coordinates across the network, presenting a severe secu…