ICML / 2026
ScaleSim: Serving Large-Scale Multi-Agent Simulation with Invocation Distance-Based Memory Management
A memory-efficient serving system for large-scale multi-agent simulation.
↗PhD student at Purdue University, building efficient systems for intelligent models.
Explore work ↓ABOUT
02I am a PhD student at Purdue University, advised by Prof. Haoran You in the EcoAI Lab. I work on systems for machine learning, with a particular interest in omni-model serving—making diverse, capable models efficient and practical at scale.
I am also interested in interactive models and world models: how models can reason, respond, and learn in rich, evolving environments.
Before Purdue, I was a research intern at UC San Diego's Picasso Lab, advised by Prof. Yufei Ding. During my undergraduate years at Zhejiang University, I worked with Prof. Bohan Zhuang in ZIP Lab and Prof. Peng Lin in CCNT Lab.
SELECTED WORK
03ICML / 2026
A memory-efficient serving system for large-scale multi-agent simulation.
↗ICML / 2026
A recurrent LIF memory module for long-horizon spiking computation.
↗NEURIPS / 2025
Workflow-aware memory management for efficient multi-agent LLM serving.
↗NOW / BEFORE
042026 — NOW
Researching efficient, scalable systems for machine learning and intelligent agents.
2025 — 2026
Built scalable and efficient inference frameworks and serving engines for multi-agent systems.
2024 — 2025
Explored computer systems and machine learning infrastructure.
2023 — 2024
Investigated power-efficient spiking neural network architectures.
FOUNDATION
05AUG 2026 — MAY 2031 (EXPECTED)
PhD in Computer Science
EcoAI Lab
West Lafayette, Indiana
2022 — 2026
B.Eng. in Computer Science
College of Computer Science and Technology
Hangzhou, China
GRADUATED 2022
High School
Beijing, China
OFF THE CLOCK
06