Keming Ye

AI4GC Lab

Keming Ye

AIGCCloud-Device CollaborationEfficient AI

About Me

Hi! I'm Keming Ye, a master's student at AI4GC Lab, Zhejiang University, advised by Prof. Shengyu Zhang.

My research focuses on AIGC acceleration, cloud-device collaboration, and efficient AI. In 2025, I interned with Tencent WXG's WeChat Vision team, where I contributed to building the UnicEdit-10M image editing dataset and benchmark. Feel free to reach out if you are interested in efficient generative AI or collaborative systems.

Research Interests

  • AIGC — Acceleration and system support for generative models.
  • Cloud-Device Collaboration — Coordinating cloud and edge resources for practical AI deployment.
  • Efficient AI — Reducing latency, compute cost, and memory overhead for modern AI workloads.

Education

  • 2024 – present · M.S. in Computer Science, Zhejiang University

Experience

  • 2025.06 – 2025.12 · Tencent WXG, WeChat Vision
    • Worked on image editing data construction and related research support.

Selected Papers

arXiv2025

UnicEdit-10M: A Dataset and Benchmark Breaking the Scale-Quality Barrier via Unified Verification for Reasoning-Enriched Edits

arXiv2026

CIAR: Interval-based Collaborative Decoding for Image Generation Acceleration