Welcome to Pervasive Intelligence Lab 👋
Join Us on Our Journey in HKUST
▼
The Pervasive Intelligence Laboratory, directed by Prof. Song Guo, is developing novel and efficient deep learning architectures, algorithms, and systems over ubiquitous mobile, IoT, and edge endpoints. By pushing the boundaries of AI capabilities, we aim to unlock new possibilities and create significant impacts in multiple areas such as scientific discovery, environmental sustainability, and societal well-being.
We focus on the following research areas:
- Cloud-Edge Collaborative Large Models
- AI Computing Cyberinfrastructure
- Trustworthy AI Governance & AIGC
- AI4Science
Join Us
We are looking for strongly motivated PhD students, Research Assistants, and Postdoctoral Fellows.
Recent Publications
(2025).
D^2MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving.
In the 31th Annual International Conference on Mobile Computing and Networking (MobiCom) (CCF-A).
(2025).
Mell: Memory-Efficient Large Language Model Serving via Multi-GPU KV Cache Management.
IEEE International Conference on Computer Communications (INFOCOM) (CCF-A).
(2025).
Causally Motivated Sycophancy Mitigation for Large Language Models.
The 13th International Conference on Learning Representations (ICLR).
(2025).
Durable Quantization Conditioned Misalignment Attack on Large Language Models.
The 13th International Conference on Learning Representations (ICLR).
(2025).
Exploring Prosocial Irrationality for LLM Agents: A Social Cognition View.
The 13th International Conference on Learning Representations (ICLR).
LAB Location
Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR
Leaflet | © OpenStreetMap contributors