Researcher

Yizeng Han

韩益增

Building efficient and adaptive models for intelligent systems. Ph.D. from Tsinghua University, advised by Prof. Gao Huang and Prof. Shiji Song.

Portrait of Yizeng Han

About Me

My research explores how generative and multimodal models can become faster, more adaptive, and more economical. I focus on efficient visual generation, dynamic neural networks, and efficient inference for visual and language models.

Recent work spans image and video diffusion, world simulation, VLM and LLM inference, and agentic model routing. I also maintain an interest in fundamental learning problems, including long-tailed learning and fine-grained learning.

Visual Generation Dynamic Neural Networks Efficient VLM Efficient Inference

Education

2018–24 Ph.D., Tsinghua UniversityAutomation
2014–18 B.E., Tsinghua UniversityAutomation

Research Experience

2024–26 Research ScientistAlibaba DAMO Academy
2023 Research Intern, MegviiFoundation Model Group · Advisor: Xiangyu Zhang
2017 Research Intern, Georgia TechAdvisor: Gregory D. Abowd

News

  • 01/2026 RAPID3 is accepted at ICLR 2026.
  • 01/2026 DyDiT++ is accepted by IEEE TPAMI.
  • 12/2025 We release a technical report and codebase for few-step text-to-image distillation.
  • 09/2025 AdaptiveNN is accepted by Nature Machine Intelligence.
  • 09/2025 FPSAttention is selected as a NeurIPS 2025 Highlight.
  • 02/2025 SGL is accepted at CVPR 2025.
  • 01/2025 DyDiT is accepted at ICLR 2025.
Earlier news
  • 11/2024 First Prize of CSIG Natural Science Award (中国图象图形学会自然科学奖一等奖).
  • 09/2024 Five works are accepted at NeurIPS 2024.
  • 07/2024 Four works are accepted at ECCV 2024.
  • 06/2024 Awarded by Outstanding Graduate of Tsinghua University, Outstanding Doctoral Dissertation, and Outstanding Graduate of Beijing.
  • 05/2024 Our work (EfficientTrain++) is accepted at TPAMI!
  • 05/2024 Our work (SimPro) is accepted at ICML 2024.
  • 04/2024 Our work (LAUDNet) is accepted at TPAMI!
  • 02/2024 Two works (GSVA and Mask Grounding) are accepted at CVPR 2024.
  • 12/2023 Our work (Learnable Semantic Data Augmentation) is accepted at TIP.
  • 10/2023 Awarded by Comprehensive Excellence Scholarship, Tsinghua University.
  • 07/2023 Three works are accepted by ICCV 2023.
  • 10/2022 Awarded by National Scholarship, Ministry of Education of China.

Selected Papers Full List

Agent-as-a-Router

Agent-as-a-Router: Agentic Model Routing for Coding Tasks

Pengfei Zhou, Zhiwei Tang, Yixing Ma, Jiasheng Tang, Yizeng Han, Zhenglin Wan, Fanqing Meng, Wei Wang, Bohan Zhuang, Wangbo Zhao, Yang You
arXiv preprint, 2026.

We formulate model routing as an agentic Context-Action-Feedback loop and introduce ACRouter and CodeRouterBench for efficient routing across frontier LLMs on coding tasks.

PDF Code
K-Forcing

K-Forcing: Joint Next-K-Token Decoding via Push-Forward Language Modeling

Zhiwei Tang, Yuanyu He, Yizeng Han, Wangbo Zhao, Jiasheng Tang, Fan Wang, Bohan Zhuang
arXiv preprint, 2026.

K-Forcing learns a push-forward language model for joint next-k-token decoding, achieving about 2.4-3.5x batch-serving speedup while reusing the autoregressive backbone.

PDF Code
MeshToken

Towards 3D-Aware Video Diffusion Models: Render-Free Human Motion Control with Mesh Tokenization

Jingyun Liang, Min Wei, Shikai Li, Yizeng Han, Hangjie Yuan, Lei Sun, Weihua Chen, Fan Wang
arXiv preprint, 2026.

We condition video diffusion models directly on compressed 3D human mesh tokens for render-free motion control with improved 3D structure and viewpoint awareness.

PDF Project
DyDiT++

DyDiT++: Diffusion Transformers with Timestep and Spatial Dynamics for Efficient Visual Generation

Wangbo Zhao*, Yizeng Han*, Jiasheng Tang, Kai Wang, Hao Luo, Yibing Song, Gao Huang, Fan Wang, Yang You
IEEE TPAMI, 2026.

We extend DyDiT to T2I (DyFLUX) and video generation. Moreover, LoRA finetuning is supported.

PDF Code
RAPID^3

RAPID^3: Tri-Level Reinforced Acceleration Policies for Diffusion Transformer

Wangbo Zhao, Yizeng Han, Zhiwei Tang, Jiasheng Tang, Pengfei Zhou, Kai Wang, Bohan Zhuang, Zhangyang Wang, Fan Wang, Yang You
ICLR, 2026.
PDF
distillation

Few-Step Distillation for Text-to-Image Generation: A Practical Guide

Yifan Pu*, Yizeng Han*, Zhiwei Tang*, Jiasheng Tang, Fan Wang, Bohan Zhuang, Gao Huang
Technical report, 2025.
PDF Code
Inferix

Inferix: A Block-Diffusion based Next-Generation Inference Engine for World Simulation

Inferix Team: Tianyu Feng, Yizeng Han, Jiahao He, Yuanyu He, Xi Lin, Teng Liu, Hanfeng Lu, Jiasheng Tang, Wei Wang, Zhiyuan Wang, Jichao Wu, Mingyang Yang, Yinghao Yu, Zeyu Zhang, Bohan Zhuang
arXiv preprint, 2025.
PDF Code
BlockVid

BlockVid: Block Diffusion for High-Quality and Consistent Minute-Long Video Generation

Zeyu Zhang, Shuning Chang, Yuanyu He, Yizeng Han, Jiasheng Tang, Fan Wang, Bohan Zhuang
arXiv preprint, 2025.
PDF Code
AdaptiveNN

Emulating Human-like Adaptive Vision for Efficient and Flexible Machine Visual Perception

Yulin Wang, Yang Yue, Yang Yue, Huanqian Wang, Haojun Jiang, Yizeng Han, Zanlin Ni, Yifan Pu, Minglei Shi, Rui Lu, Qisen Yang, Andrew Zhao, Zhuofan Xia, Shiji Song, Gao Huang
Nature Machine Intelligence, 2025.
PDF Code
FPSAttention

FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion

Akide Liu, Zeyu Zhang, Zhexin Li, Xuehai Bai, Yizeng Han, Jiasheng Tang, Yuanjie Xing, Jichao Wu, Mingyang Yang, Weihua Chen, Jiahao He, Yuanyu He, Fan Wang, Gholamreza Haffari, Bohan Zhuang
NeurIPS (Highlight), 2025.
PDF Project
SGL

A Stitch in Time Saves Nine: Small VLM is a Precise Guidance for accelerating Large VLMs

Wangbo Zhao*, Yizeng Han*, Jiasheng Tang, Zhikai Li, Yibing Song, Kai Wang, Zhangyang Wang, Yang You
CVPR, 2025.
PDF Code
DyT

Dynamic Tuning Towards Parameter and Inference Efficiency for ViT Adaptation

Wangbo Zhao, Jiasheng Tang, Yizeng Han, Yibing Song, Kai Wang, Gao Huang, Fan Wang, Yang You
NeurIPS, 2024.
PDF Code
Deer

DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution

Yang Yue, Yulin Wang, Bingyi Kang, Yizeng Han, Shenzhi Wang, Shiji Song, Jiashi Feng, Gao Huang
NeurIPS, 2024.
PDF Code
Survey

Dynamic Neural Networks: A Survey

Yizeng Han*, Gao Huang*, Shiji Song, Le Yang, Honghui Wang, Yulin Wang
IEEE TPAMI, 2021.

In this survey, we comprehensively review the rapidly developing area, dynamic neural networks.

PDF 智源社区 Slides
DyDiT++

DyDiT++: Diffusion Transformers with Timestep and Spatial Dynamics for Efficient Visual Generation

Wangbo Zhao*, Yizeng Han*, Jiasheng Tang, Kai Wang, Hao Luo, Yibing Song, Gao Huang, Fan Wang, Yang You
IEEE TPAMI, 2026.

We extend DyDiT to T2I (DyFLUX) and video generation. Moreover, LoRA finetuning is supported.

PDF Code
LAUDNet

Latency-aware Unified Dynamic Networks for Efficient Image Recognition

Yizeng Han*, Zeyu Liu*, Zhihang Yuan*, Yifan Pu, Chaofei Wang, Shiji Song, Gao Huang
IEEE TPAMI, 2024.
PDF Code
SGL

A Stitch in Time Saves Nine: Small VLM is a Precise Guidance for accelerating Large VLMs

Wangbo Zhao*, Yizeng Han*, Jiasheng Tang, Zhikai Li, Yibing Song, Kai Wang, Zhangyang Wang, Yang You
CVPR, 2025.
PDF Code
DyDiT

Dynamic Diffusion Transformer

Wangbo Zhao, Yizeng Han, Jiasheng Tang, Kai Wang, Yibing Song, Gao Huang, Fan Wang, Yang You
ICLR, 2025.
PDF Code
Dyn_Perceiver

Dynamic Perceiver for Efficient Visual Recognition

Yizeng Han*, Dongchen Han*, Zeyu Liu, Yulin Wang, Xuran Pan, Yifan Pu, Chao Deng, Junlan Feng, Shiji Song, Gao Huang
ICCV, 2023.
PDF Code
L2W-DEN

Learning to Weight Samples for Dynamic Early-exiting Networks

Yizeng Han*, Yifan Pu*, Zihang Lai, Chaofei Wang, Shiji Song, Junfen Cao, Wenhui Huang, Chao Deng, Gao Huang
ECCV, 2022.
PDF Code
LASNet

Latency-aware Spatial-wise Dynamic Networks

Yizeng Han*, Zhihang Yuan*, Yifan Pu*, Chenhao Xue, Shiji Song, Guangyu Sun, Gao Huang
NeurIPS, 2022.
PDF Code
RANet

Resolution Adaptive Networks for Efficient Inference

Le Yang*, Yizeng Han*, Xi Chen*, Shiji Song, Jifeng Dai, Gao Huang
CVPR, 2020.
PDF Code
SAR

Spatially Adaptive Feature Refinement for Efficient Inference

Yizeng Han, Gao Huang, Shiji Song, Le Yang, Yitian Zhang, Haojun Jiang
IEEE TIP, 2021.
PDF
SimPro

SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning

Chaoqun Du*, Yizeng Han*, Gao Huang
ICML, 2024.
PDF Code
LearnableISDA

Fine-grained Recognition with Learnable Semantic Data Augmentation

Yifan Pu*, Yizeng Han*, Yulin Wang, Junlan Feng, Chao Deng, Gao Huang
IEEE TIP, 2023.
PDF Code

Awards

  • 2024First Prize, CSIG Natural Science Award · 中国图象图形学会自然科学奖一等奖（第四完成人）
  • 2024Outstanding Doctoral Dissertation · Tsinghua University
  • 2024Outstanding Graduate · Tsinghua University
  • 2024Outstanding Graduate of Beijing · Beijing Municipal Education Commission
  • 2023Comprehensive Excellence Scholarship · Tsinghua University
  • 2022National Scholarship · Ministry of Education of China
Earlier honors
  • 2017Comprehensive Excellence Scholarship · Tsinghua University
  • 2016Comprehensive Excellence Scholarship · Tsinghua University
  • 2015Academic Excellence Scholarship · Tsinghua University

Get in Touch

For research discussions and collaborations, feel free to reach out.

yizeng38@gmail.com