Hancheng Ye

Ph.D. Student in Duke University

“Welcome to my world!”

About me:

  • Hi, I am a second year PhD student at Duke ECE, advised by Prof. Yiran Chen. Previously, I obtained my M.S. and B.E. from School of Information Science and Technology, Fudan University, where I worked closely with Prof. Tao Chen and Dr. Bo Zhang (Shanghai Artificial Intelligence Laboratory) in the area of efficient deep learning (specifically, model compression for multi-task learning). My current research interests lie mainly in Machine Learning System, where I work closely with Prof. Danyang Zhuo on the efficient multi-agent system design. In the summer before coming to Duke, I was very lucky to work with Dr. Zhijian Liu on the efficient diffusion acceleration and Dr. Bo Zhang on multimodal learning. I interned at Shanghai Artificial Intelligence Laboratory and MIT HAN Lab.
  • For more information, please click here.
  • Social media: Google Scholar · LinkedIn · X · GitHub

News

  • Oct 2025 — We opensource the KVCOMM codebase. See Paper and X Blog.
  • Sep 2025 — Two papers are accepted to NeurIPS 2025 (KVCOMM and GAINRL). Congrats to Qinsi!
  • Aug 2025 — ChartX & ChartVLM is accepted to T-IP 2025. Congrats to Renqiu!
  • May 2025 — Two papers (SADA and CoreMatching) are accepted to ICML 2025. Congrats to Justin, Yixiao, and Qinsi!
  • Feb 2025 — BridgeNet is accepted to T-PAMI 2025. Congrats to Jingdong!
  • Jan 2025 — GeoX is accepted to ICLR 2025. Congrats to Renqiu and Mingsheng!
  • Oct 2024 — AdaptiveDiffusion is accepted to NeurIPS 2024, BDP-DARTS is accepted to T-CSVT 2024, and one paper is accepted to Neurocomputing 2024. Congrats to Chongjun and Peng!
  • Aug 2024 — Start my Ph.D. journey in Duke University!
  • Jul 2024 — SPOT is accepted to T-PAMI 2024. Congrats to Xiangchao and Runjian!
  • Mar 2024 — OFB is accepted to CVPR 2024, and STP is accepted to ECCV 2024. Congrats to Shengji and Weihao!
  • Mar 2023 — PAGCP is accepted to T-PAMI 2023.

Highlights

KVCOMM
NeurIPS 2025 — KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems Authors: Hancheng Ye, Zhengqi Gao, Mingyuan Ma, Qinsi Wang, Yuzhe Fu, Ming-Yu Chung, Yueqian Lin, Zhijian Liu, Jianyi Zhang, Danyang Zhuo, Yiran Chen. GitHub Paper
ChartX & ChartVLM
T-IP 2025 — ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning Authors: Renqiu Xia*, Hancheng Ye*, Xiangchao Yan, Qi Liu, Hongbin Zhou, Zijun Chen, Botian Shi, Junchi Yan, and Bo Zhang. GitHub Paper
SADA: Stability-guided Adaptive Diffusion Acceleration
ICML 2025 — SADA: Stability-guided Adaptive Diffusion Acceleration Authors: Ting Jiang*, Yixiao Wang*, Hancheng Ye*, Zishan Shao, Jingwei Sun, Jingyang Zhang, Zekai Chen, Jianyi Zhang, Yiran Chen, Hai Li. GitHub Paper
Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy
NeurIPS 2024 — Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy Authors: Hancheng Ye*, Jiakang Yuan*, Renqiu Xia, Xiangchao Yan, Tao Chen, Junchi Yan, Botian Shi, Bo Zhang. GitHub Paper
OFB: Once-for-Both (CVPR 2024)
CVPR 2024 — Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression Authors: Hancheng Ye, Chong Yu, Peng Ye, Renqiu Xia, Yansong Tang, Jiwen Lu, Tao Chen, Bo Zhang. GitHub Paper
Performance-aware Approximation of Global Channel Pruning for Multitask CNNs
T-PAMI 2023 — Performance-aware Approximation of Global Channel Pruning for Multitask CNNs Authors: Hancheng Ye, Bo Zhang, Tao Chen, Jiayuan Fan, and Bin Wang. GitHub Paper

Selected Publications

Efficient Machine Learning

  • KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems. NeurIPS 2025. Hancheng Ye, Zhengqi Gao, Mingyuan Ma, Qinsi Wang, Yuzhe Fu, Ming-Yu Chung, Yueqian Lin, Zhijian Liu, Jianyi Zhang, Danyang Zhuo, Yiran Chen.
  • SADA: Stability-guided Adaptive Diffusion Acceleration. ICML 2025. Ting Jiang*, Yixiao Wang*, Hancheng Ye*, Zishan Shao, Jingwei Sun, Jingyang Zhang, Zekai Chen, Jianyi Zhang, Yiran Chen, Hai Li.
  • Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy. NeurIPS 2024. Hancheng Ye*, Jiakang Yuan*, Renqiu Xia, Xiangchao Yan, Tao Chen, Junchi Yan, Botian Shi, Bo Zhang.
  • Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression. CVPR 2024. Hancheng Ye, Chong Yu, Peng Ye, Renqiu Xia, Yansong Tang, Jiwen Lu, Tao Chen, Bo Zhang.
  • PAGCP: Performance-aware Approximation of Global Channel Pruning for Multitask CNNs. T-PAMI 2023. Hancheng Ye, Bo Zhang, Tao Chen, Jiayuan Fan, Bin Wang.
  • Angles don't lie: Unlocking training-efficient RL through the model's own signals. NeurIPS 2025 (Spotlight). Qinsi Wang, Jinhan Ke, Hancheng Ye, Yueqian Lin, Yuzhe Fu, Jianyi Zhang, Kurt Keutzer, Chenfeng Xu, Yiran Chen.
  • Efficient architecture search via bi-level data pruning. T-CSVT 2024. Chongjun Tu, Peng Ye, Weihao Lin, Hancheng Ye, Chong Yu, Tao Chen, Baopu Li, & Wanli Ouyang.
  • Enhanced sparsification via stimulative training. ECCV 2024. Shengji Tang*, Weihao Lin*, Hancheng Ye, Peng Ye, Chong Yu, Baopu Li, Tao Chen.
  • Sample-centric feature generation for semi-supervised few-shot learning. T-IP 2022. Bo Zhang, Hancheng Ye, Gang Yu, Bin Wang, Yike Wu, Jiayuan Fan, Tao Chen.

Multimodal Learning

  • ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning. T-IP 2025. Renqiu Xia*, Hancheng Ye*, Xiangchao Yan, Qi Liu, Hongbin Zhou, Zijun Chen, Botian Shi, Junchi Yan, Bo Zhang.
  • GeoX: Geometric Problem Solving Through Unified Formalized Vision-Language Pre-training. ICLR 2025. Renqiu Xia*, Mingsheng Li*, Hancheng Ye, Wenjie Wu, Hongbin Zhou, Jiakang Yuan, Tianshuo Peng, Xinyu Cai, Xiangchao Yan, Bin Wang, Conghui He, Botian Shi, Tao Chen, Junchi Yan, Bo Zhang.

Contact Me