One Paper Accepted by TCAD'2026 and One Paper Accepted by CoRL'2026
One paper on efficient MoE inference is accepted by TCAD'2026 and one paper on efficient VLA is accepted by CoRL'2026.
The title of the accepted TCAD paper is “HDA-MoE: Hybrid Parallelism and Dynamic, Adaptive Scheduling for Mixture-of-Experts with 3D Near-Memory Processing”, featuring a novel scheduling algorithm for MoE inference on 3D DRAM accelerators. This is a journal extension of our HD-MoE paper accepted by ICCAD'2025. The work is led by our PhD student Haochen Huang.
The title of the accepted CoRL paper is “Jetson-PI: Towards Onboard Real-Time Robot Control via Foresight-Aligned Asynchronous Inference”, featuring asynchronous inference optimization and edge GPU deployment optimization. The work is led by our PhD student Zebin Yang.