Jayce Xu

Huazhong University of Science and Technology

Jayce Xu

徐晟杰

Hi, I’m Jayce Xu (徐晟杰). I am a third-year undergraduate in Software Engineering at Huazhong University of Science and Technology. I currently intern in the Antibody Design Group, Foundational Algorithms Department at BioMap (百图生科), and previously interned as an AI drug-discovery algorithm researcher at Baidu PaddlePaddle (Helix Team). I was awarded the National Scholarship of China (2024–2025) and the HUST Youth May 4th Medal (2026) based on comprehensive academic performance and research output. Despite being an undergraduate, I have been actively engaged in AI for Science and Medicine research, focusing on:

AI drug discovery
Multi-agent systems
Graph neural networks

I’m also interested in model security (adversarial attacks, fake detection). As an undergraduate researcher, I am actively seeking PhD opportunities to further advance my research career.

Contact:

Email: shengjiexu99@gmail.com / u202317280@hust.edu.cn
WeChat: Jayce_Xu99

If you are a professor or collaborator working on AI for science/medicine, security, or scalable agent systems, I would be deeply honored to connect and learn about potential opportunities.

Publications

Dumpling GNN: Hybrid GNN Enables Better ADC Payload Activity Prediction Based on the Chemical Structure. Xu, S.; Xie, L.; Dai, R.; Lyu, Z.Int. J. Mol. Sci. 2025, 26, 4859. https://doi.org/10.3390/ijms26104859
First author & corresponding author
FastJSMA: Accelerating Jacobian-based Saliency Map Attacks through Gradient Decoupling — Zhenghao Gao, Shengjie Xu, Zijing Li, Meixi Chen, Chaojian Yu, Yuanjie Shao, Changxin Gao. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025, pp. 1506-1515. ICCV 2025
Co-first author
Season: Spectrum-Aware Orthogonal Gradient Refinement for Transfer-Based Adversarial Attacks — Tianyi Wang, Zhenghao Gao, Shengjie Xu. Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), 2026. (accepted March 17, 2026; CCF-B; Spotlight presentation) Publications page
Corresponding author
End-to-End Reverse Screening Identifies Protein Targets of Small Molecules Using HelixFold3 — Shengjie Xu, Xianbin Ye, Mengran Zhu, Xiaonan Zhang, Shanzhuo Zhang, Xiaomin Fang. arXiv:2601.13693, 2026. (accepted pending major revision at IEEE JBHI) arXiv
First author

Work Experience

BioMap (百图生科) — Algorithm Intern, Antibody Design Group, Foundational Algorithms Dept. (Jun 2026–Present) Antibody design algorithms in the Foundational Algorithms Department
VLR Lab, Huazhong University of Science and Technology — Group Lead, Drug Repurposing (Dec 2023–Jun 2025) Research on applying multi‑agent systems and graph neural networks to drug repurposing
Hangzhou Intelligence‑Driven Education Technology — Deputy Director, Big Data Algorithms (Aug 2024–Mar 2025) Application‑oriented knowledge graphs and agent systems for education
Guangzhou Institutes of Biomedicine and Health, CAS — Assistant Researcher (Mar 2025–Aug 2025) AI for spatial transcriptomics and drug sensitivity analysis
Huazhong University of Science and Technology — Teaching Assistant, Data Intelligence Analysis & Visualization (Mar 2025–Aug 2025) Teaching assistance for data intelligence analysis and visualization courses, leading hands-on practice sessions and guiding students in data mining competitions
Baidu PaddlePaddle (Helix) — Algorithm Intern (Jun 2025–Jan 2026) AI‑assisted drug screening; co-authored HelixFold3 reverse screening paper (arXiv:2601.13693, accepted pending major revision at IEEE JBHI)

Competitions

HUST Youth May 4th Medal (student category; 10 recipients: 8 postgraduates + 2 undergraduates; Jayce was one of two undergraduates) (May 2026)
19th “Challenge Cup” National Undergraduate Extracurricular Academic Science and Technology Works Competition “AI+” Special Track — Special Prize (OmnMedical project; topic: AI-based intelligent agent construction for drugs and advanced drug delivery)
Qiushi Cup Undergraduate Research Competition — University First Prize, individual project (Mar 2025)
Global Campus AI Algorithm Elite Challenge — National Second Prize, team leader (Dec 2024)
Challenge Cup 2024 “List to Solve” Special Track — National First Prize, project lead (Nov 2024)
HUST 10th Mobile App Design Contest — Programming Group Champion (Mar 2024)
CCF BDCI Ernie Bot Plugin Track — National Third Prize (Nov 2023) </small>

Hobbies & Interests

Sports: Badminton
Gaming: MOBA, Fighting games, ARPG
Research: Mythology and folklore studies
Music: Singing

Life snapshots

selected publications

Int. J. Mol. Sci.
Dumpling GNN: Hybrid GNN Enables Better ADC Payload Activity Prediction Based on the Chemical Structure

Shengjie Xu, Linyi Xie, Rui Dai, and 1 more author

International Journal of Molecular Sciences, May 2025

Abs DOI Bib HTML

Antibody-drug conjugates (ADCs) represent a promising class of targeted cancer therapeutics that combine the specificity of monoclonal antibodies with the cytotoxicity of small-molecule drugs. The payload activity prediction is crucial for ADC design and optimization. In this study, we propose Dumpling GNN, a hybrid graph neural network architecture that leverages both molecular graph structure and chemical features to predict ADC payload activity. Our approach integrates multiple graph neural network layers with attention mechanisms to capture complex molecular interactions and structural patterns. Experimental results demonstrate that Dumpling GNN significantly outperforms traditional machine learning methods and existing graph neural network approaches in predicting ADC payload activity based on chemical structure. The model achieves improved accuracy in identifying potent payload candidates, which can accelerate the drug discovery process for ADCs.
@article{xu2025dumplinggnn, title = {Dumpling GNN: Hybrid GNN Enables Better ADC Payload Activity Prediction Based on the Chemical Structure}, author = {Xu, Shengjie and Xie, Linyi and Dai, Rui and Lyu, Ziyi}, journal = {International Journal of Molecular Sciences}, volume = {26}, number = {10}, pages = {4859}, year = {2025}, month = may, publisher = {MDPI}, doi = {10.3390/ijms26104859}, url = {https://www.mdpi.com/1422-0067/26/10/4859}, }

ICCV

FastJSMA: Accelerating Jacobian-based Saliency Map Attacks through Gradient Decoupling

Zhenghao Gao, Shengjie Xu, Zijing Li, and 4 more authors

In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Oct 2025

Abs Bib HTML

FastJSMA is an efficient and interpretable adversarial attack that introduces a gradient decoupling mechanism to decompose Jacobian-based saliency into complementary suppression and excitation gradients, reducing the computational complexity of JSMA while maintaining high attack success rates on large-scale datasets.

@inproceedings{gao2025fastjsma,
  title = {FastJSMA: Accelerating Jacobian-based Saliency Map Attacks through Gradient Decoupling},
  author = {Gao, Zhenghao and Xu, Shengjie and Li, Zijing and Chen, Meixi and Yu, Chaojian and Shao, Yuanjie and Gao, Changxin},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  year = {2025},
  month = oct,
  pages = {1506--1515},
  url = {https://openaccess.thecvf.com/content/ICCV2025/html/Gao_FastJSMA_Accelerating_Jacobian-based_Saliency_Map_Attacks_through_Gradient_Decoupling_ICCV_2025_paper.html},
}

ICME
Season: Spectrum-Aware Orthogonal Gradient Refinement for Transfer-Based Adversarial Attacks

Tianyi Wang, Zhenghao Gao, and Shengjie Xu

In Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), Oct 2026

Accepted March 17, 2026. CCF-B. Spotlight presentation.

Spotlight Abs Bib

Selected for Spotlight presentation at ICME 2026.

Transfer-based adversarial attacks often transfer poorly across heterogeneous architectures because CNNs favor local textures while Vision Transformers rely on global shapes. Season decomposes each update into a low-frequency structural branch and a high-frequency textural branch, uses low-saliency guidance to steer high-frequency energy toward background regions, and applies an orthogonal projection so textural updates lie in the orthogonal complement of the structural direction. As a training-free plug-and-play wrapper, it improves transfer success on ImageNet against diverse CNN, ViT, and MLP targets.
@inproceedings{wang2026season, title = {Season: Spectrum-Aware Orthogonal Gradient Refinement for Transfer-Based Adversarial Attacks}, author = {Wang, Tianyi and Gao, Zhenghao and Xu, Shengjie}, booktitle = {Proceedings of the IEEE International Conference on Multimedia and Expo (ICME)}, year = {2026}, note = {Accepted March 17, 2026. CCF-B. Spotlight presentation.}, }
arXiv
End-to-End Reverse Screening Identifies Protein Targets of Small Molecules Using HelixFold3

Shengjie Xu, Xianbin Ye, Mengran Zhu, and 3 more authors

arXiv preprint arXiv:2601.13693, Oct 2026

Accepted pending major revision at IEEE Journal of Biomedical and Health Informatics (JBHI)

Abs Bib HTML

We present an end-to-end reverse screening pipeline that leverages HelixFold3 to jointly fold proteins and dock small-molecule ligands, enabling large-scale identification of protein targets for small molecules without separate structure preparation. Validated across approximately 100 small molecules, our pipeline outperforms conventional reverse docking workflows in both hit rate and computational efficiency.
@article{xu2026helixfold3, title = {End-to-End Reverse Screening Identifies Protein Targets of Small Molecules Using HelixFold3}, author = {Xu, Shengjie and Ye, Xianbin and Zhu, Mengran and Zhang, Xiaonan and Zhang, Shanzhuo and Fang, Xiaomin}, journal = {arXiv preprint arXiv:2601.13693}, year = {2026}, note = {Accepted pending major revision at IEEE Journal of Biomedical and Health Informatics (JBHI)}, url = {https://arxiv.org/abs/2601.13693}, }

Watertox: The Art of Simplicity in Universal Attacks A Cross-Model Framework for Robust Adversarial Generation

Zhenghao Gao, Shengjie Xu, Meixi Chen, and 1 more author

Oct 2024

Bib

@misc{gao2024watertoxartsimplicityuniversal,
  title = {Watertox: The Art of Simplicity in Universal Attacks A Cross-Model Framework for Robust Adversarial Generation},
  author = {Gao, Zhenghao and Xu, Shengjie and Chen, Meixi and Zhao, Fangyao},
  year = {2024},
  eprint = {2412.15924},
  archiveprefix = {arXiv},
  primaryclass = {cs.CV},
  url = {https://arxiv.org/abs/2412.15924},
}