Hongyang Jia

Tsinghua University

Papers

1

Total Citations

1

H-Index

1

About

Hongyang Jia is a leading researcher at the frontier of energy-efficient edge AI and robotic intelligence. His work centers on enabling real-time, on-device machine learning for autonomous systems, with a particular focus on diffusion-based models for robot manipulation. Jia’s major contribution is the development of a novel edge system-on-chip (SoC) that achieves both low-latency inference and high-fidelity on-device fine-tuning—a breakthrough that overcomes the traditional trade-off between speed and adaptability. His most-cited paper, "A 94Hz Inference and 7.4mJ/Epoch Fine-Tune Edge SoC for Diffusion-Based Robot Manipulation with Speculation and Disturbance Enhancement," introduces a speculative parallel inference technique that dramatically accelerates diffusion transformer-based action generation. This work, with 1 citation, represents a foundational step toward practical, energy-efficient robotic control at the edge. Jia’s research is pivotal for deploying sophisticated AI in resource-constrained environments, promising to unlock new capabilities in autonomous robotics, from industrial automation to assistive technologies. His achievements mark him as a rising star in the intersection of hardware design and embodied AI.

Research Focus

Key Achievements

1
H-Index
1
Papers
1
Total Citations
1
Avg Citations/Paper
🏆 Most Cited Paper
A 94Hz Inference and 7.4mJ/Epoch Fine-Tune Edge SoC for Diffusion-Based Robot Manipulation with Speculation and Disturbance Enhancement
1 citations · 2025
📈 Most Prolific Year: 2025 (1 Papers)
🤝 Key Collaborators: 4
🏛 Institutions: Tsinghua University

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago