Zhongzhan Huang (黄中展)

Hi, I am Zhongzhan Huang (黄中展), a Final-year Computer Science PhD student at Sun Yat-sen University, advised by Prof. Liang Lin. My research interests include the foundations of deep learning (e.g. network architectures, attention mechanisms, LLMs and diffusion models), and recommender system.

A funny finding : $2^{2^{2}} + 2^{2} + 2 = 22$

Academic Service

Conference Reviewer: NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, EMNLP, ACL, AAAI, WWW, MM
Journal Reviewer: IEEE TMM, IEEE TCSVT, IEEE TNNLS
Area Chair: BMVC

I plan to search for a full-time research/applied scientist/postdoc position. Please feel free to contact me at huangzhzh23@mail2.sysu.edu.cn!

Contact: zhongzhanhuang [at] foxmail [dot] com
Follow: Google Scholar | github

News

Mar, 2025	We release our RouterEval for promising Routing LLM paradigm [paper] [slide] [code] [resource].
Feb, 2025	Cheers! LoTbench, accepted to IEEE TPAMI and DVIB, accepted to WWW’25 !
Jul, 2024	Cheers! ASR, accepted to ECCV’24 !
May, 2024	Cheers! AttNS, accepted to ICML’24 ! and LTH4SA, accepted to ICME’24 !
Feb, 2024	Cheers! Mirror Gradient, accepted to WWW’24 ! and CLoT, accepted to CVPR’24 !
Sep, 2023	Cheers! NeurVec, accepted to Sci. Reports ! and ScaleLong, accepted to NeurIPS’23 !
Jul, 2023	Cheers! StepNet, accepted to ICCV’23 and SUR-adapter, accepted to ACM MM’23 (Oral) !
Jan, 2023	I join Sea AI Lab as a research intern supervised by Prof. Pan Zhou and Prof. Shuicheng Yan, Enjoy it!
Oct, 2022	Cheers! CEM, accepted to EMNLP’22 !
Jan, 2022	Cheers! SANN, accepted to ICLR’22 !
Sep, 2021	Cheers! CWDA, accepted to NeurIPS’21 !
Sep, 2021	I go back to the campus and start pursuing PhD degree in SYSU! Enjoy it!

Selected Publications (click here for all)

Arxiv

Routereval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMs

Zhongzhan Huang, Guoming Ling, Vincent Liang, Yupei Lin, Yandong Chen, Shanshan Zhong, Hefeng Wu, and Liang Lin

In arXiv preprint (in submission), preprint

PDF Code

@inproceedings{router,
  title = {Routereval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMs},
  author = {Huang, Zhongzhan and Ling, Guoming and Liang, Vincent and Lin, Yupei and Chen, Yandong and Zhong, Shanshan and Wu, Hefeng and Lin, Liang},
  booktitle = {arXiv preprint (in submission)},
  year = {preprint}
}

IEEE TPAMI

A Causality-aware Paradigm for Evaluating Creativity of Multimodal Large Language Models

Zhongzhan Huang, Shanshan Zhong, Pan Zhou, Shanghua Gao, Zitnik Marinka, and Liang Lin

In IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI), 2025

PDF Code

@inproceedings{pami,
  title = {A Causality-aware Paradigm for Evaluating Creativity of Multimodal Large Language Models},
  author = {Huang, Zhongzhan and Zhong, Shanshan and Zhou, Pan and Gao, Shanghua and Marinka, Zitnik and Lin, Liang},
  booktitle = {IEEE Transactions on Pattern Analysis and Machine Intelligence (<strong>IEEE TPAMI</strong>)},
  year = {2025}
}

ECCV’24

ASR: Stripe Observation Guided Inference Cost-free Attention Mechanism

Zhongzhan Huang, Shanshan Zhong, Wushao Wen, Jinghui Qin, and Liang Lin

In The European Conference on Computer Vision (ECCV), 2024

PDF Code

@inproceedings{asr,
  title = {ASR: Stripe Observation Guided Inference Cost-free Attention Mechanism},
  author = {Huang, Zhongzhan and Zhong, Shanshan and Wen, Wushao and Qin, Jinghui and Lin, Liang},
  booktitle = {The European Conference on Computer Vision (<strong>ECCV</strong>)},
  year = {2024}
}

ICML’24

AttNS: Attention-Inspired Numerical Solving For Limited Data Scenarios

Zhongzhan Huang, Mingfu Liang, Shanshan Zhong, and Liang Lin

In The International Conference on Machine Learning (ICML), 2024

PDF Code

@inproceedings{AttSolver1,
  title = {AttNS: Attention-Inspired Numerical Solving For Limited Data Scenarios},
  author = {Huang, Zhongzhan and Liang, Mingfu and Zhong, Shanshan and Lin, Liang},
  booktitle = {The International Conference on Machine Learning (<strong>ICML</strong>)},
  year = {2024}
}

CVPR’24

Let’s Think Outside the Box: Exploring Leap-of-Thought in LLMs with Creative Humor Generation

Shanshan Zhong, (co-first) Zhongzhan Huang, Shanghua Gao, Wushao Wen, Liang Lin, Marinka Zitnik, and Pan Zhou

In The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

PDF Code

@inproceedings{scale,
  title = {Let’s Think Outside the Box: Exploring Leap-of-Thought in LLMs with Creative Humor Generation},
  author = {Zhong, Shanshan and Huang, (co-first) Zhongzhan and Gao, Shanghua and Wen, Wushao and Lin, Liang and Zitnik, Marinka and Zhou, Pan},
  booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (<strong>CVPR</strong>)},
  year = {2024},
}

NeurIPS’23

ScaleLong: Towards More Stable Training of Diffusion Model via Scaling Network Long Skip Connection

Zhongzhan Huang, Pan Zhou, Shuicheng Yan, and Liang Lin

In The Conference on Neural Information Processing Systems (NeurIPS), 2023

PDF Code

@inproceedings{scalf,
  title = {ScaleLong: Towards More Stable Training of Diffusion Model via Scaling Network Long Skip Connection},
  author = {Huang, Zhongzhan and Zhou, Pan and Yan, Shuicheng and Lin, Liang},
  booktitle = {The Conference on Neural Information Processing Systems (<strong>NeurIPS</strong>)},
  year = {2023},
}

WWW’24

Mirror Gradient: Towards Robust Multimodal Recommender Systems via Exploring Flat Local Minima

Shanshan Zhong, (co-first) Zhongzhan Huang, Daifeng Li, Wushao Wen, Jinghui Qin, and Liang Lin

In International World Wide Web Conference (WWW), 2024

PDF Code

@inproceedings{mg,
  title = {Mirror Gradient: Towards Robust Multimodal Recommender Systems via Exploring Flat Local Minima},
  author = {Zhong, Shanshan and Huang, (co-first) Zhongzhan and Li, Daifeng and Wen, Wushao and Qin, Jinghui and Lin, Liang},
  booktitle = {International World Wide Web Conference  (<strong>WWW</strong>)},
  year = {2024}
}

Sci. Reports

On Fast Simulation of Dynamical System with Neural Vector Enhanced Numerical Solver

Zhongzhan Huang, Senwei Liang, Hong Zhang, Haizhao Yang, and Liang Lin

In Scientific Reports (Sci. Reports), 2023

arXiv Code

@inproceedings{huang2022accelerating,
  title = {On Fast Simulation of Dynamical System with Neural Vector Enhanced Numerical Solver},
  author = {Huang, Zhongzhan and Liang, Senwei and Zhang, Hong and Yang, Haizhao and Lin, Liang},
  booktitle = {Scientific Reports (<strong>Sci. Reports</strong>)},
  year = {2023}
}

ACM MM’23

SUR-adapter: Enhancing text-to-image pre-trained diffusion models with large language models

Shanshan Zhong, (co-first) Zhongzhan Huang, Wushao Wen, Jinghui Qin, and Liang Lin

In 31th ACM International Conference on Multimedia (ACM MM Oral), 2023

arXiv Code

@inproceedings{zhong2023adapter,
  title = {SUR-adapter: Enhancing text-to-image pre-trained diffusion models with large language models},
  author = {Zhong, Shanshan and Huang, (co-first) Zhongzhan and Wen, Wushao and Qin, Jinghui and Lin, Liang},
  booktitle = {31th ACM International Conference on Multimedia (<strong>ACM MM </strong> <strong style="color:red;">Oral</strong>)},
  year = {2023},
  organization = {ACM Press}
}

ICCV’23

Understanding Self-attention Mechanism via Dynamical System Perspective

Zhongzhan Huang, Mingfu Liang, Jinghui Qin, Shanshan Zhong, and Liang Lin

In The IEEE/CVF International Conference on Computer Vision (ICCV), 2023

arXiv

@inproceedings{huang2023understanding,
  title = {Understanding Self-attention Mechanism via Dynamical System Perspective},
  author = {Huang, Zhongzhan and Liang, Mingfu and Qin, Jinghui and Zhong, Shanshan and Lin, Liang},
  booktitle = {The IEEE/CVF International Conference on Computer Vision (<strong>ICCV</strong>)},
  year = {2023}
}

ICLR’22

Stiffness-aware neural network for learning Hamiltonian systems

Senwei Liang, Zhongzhan Huang, and Hong Zhang

In International Conference on Learning Representations (ICLR), 2022

PDF Code

@inproceedings{liang2021stiffness,
  title = {Stiffness-aware neural network for learning Hamiltonian systems},
  author = {Liang, Senwei and Huang, Zhongzhan and Zhang, Hong},
  booktitle = {International Conference on Learning Representations (<strong>ICLR</strong>)},
  year = {2022}
}

NeurIPS’21

Rethinking the Pruning Criteria for Convolutional Neural Network

Zhongzhan Huang, Wenqi Shao, Xinjiang Wang, Liang Lin, and Ping Luo

In The Conference on Neural Information Processing Systems (NeurIPS), 2021

PDF

@inproceedings{huang2021convolution,
  title = {Rethinking the Pruning Criteria for Convolutional Neural Network},
  author = {Huang, Zhongzhan and Shao, Wenqi and Wang, Xinjiang and Lin, Liang and Luo, Ping},
  booktitle = {The Conference on Neural Information Processing Systems (<strong>NeurIPS</strong>)},
  year = {2021},
}

This page has been accessed at least

times since 10/10/2021, and on average

per day.