Haodong Zhao 赵皓东

Biography

I am a fourth-year PhD candidate at School of Computer Science, Shanghai Jiao Tong University (SJTU), supervised by Prof. Gongshen Liu and Prof. Zhuosheng Zhang. I am also supervised by Prof. Bingsheng He at School of Computing, National University of Singapore (NUS) and collaborate with Dr. Zhaomin Wu.

I got my bachelor's degree from School of Electronic Information and Electrical Engineering at Shanghai Jiao Tong University in 2021, supervised by Prof. Gongshen Liu.

My general research interest lies in LLM Agent, AI Safety/Security/Privacy and Trustworthy Distributed Learning.

Prospective collaborators: I am open to collaborate. If you're interested in working with me, feel free to send me an email. I will join NUS as a Research Fellow (Postdoc) at the end of 2026. 我计划在26年12月前往NUS做博士后，欢迎对我的研究感兴趣的同学邮件/微信联系我合作，尤其欢迎计划选择我在SJTU/NUS的导师的本科生/研究生。

Education

• 2024 - 2026, visiting PhD Student, School of Computing (SoC), National University of Singapore
• 2021 - now, PhD Student, School of Computer Science (SCS), Shanghai Jiao Tong University
• 2017 - 2021, Bachelor, School of Electronic Information and Electrical Engineering (SEIEE), Shanghai Jiao Tong University

News

Affiliation

• Tongyi Lab
  Alibaba Group
• AI Security Lab (Unicorn), belongs to Institute of Artificial Intelligence and Cognitive Security
  School of Computer Science, Shanghai Jiao Tong University
• Xtra Computing Group
  School of Computing, National University of Singapore

Internship

• 2026.6 - now, Alibaba AIR project, Research Intern (Tongyi Lab, MAS), Alibaba
• 2024.6 - 2025.6, Tencent Rhino-bird Research Elite Program(腾讯犀牛鸟精英人才计划). Research Intern (WXG, FedLLM)

Honors & Awards

• 2026 Outstanding Graduate of Shanghai
• 2025 The CAST Youth Science and Technology Talent Cultivation Program (PhD, 中国科协青年科技人才培育工程博士生专项计划)
• 2023 Outstanding Teaching Assistant of Shanghai Jiaotong University
• 2021 Yang Yuanqing Education Fund Outstanding Undergraduate Excellence Scholarship
• 2021 Outstanding Graduate of Shanghai
• 2021 Rongchang Leadership Scholarship

Tutorials and Contributions

《大模型开发全流程》

Organizing and participating.

Course Link

Dive into LLMs《动手学大模型》

Participating.

Course Link

Paper Rebuttal Tips

Organizing and participating (MLNLP community).

Link

SecretFlow

Contributors.

GitHub Community

LLM-DNA

Extract LLM DNA vectors — low-dimensional representations of model.

PyPI

UniDial-EvalKit

A unified framework focusing on the evaluation of multi-turn interactions in LLMs.

GitHub

Selected Publications

Full publication list available on Google Scholar. (* denotes equal contribution, † denotes corresponding author)

Preprint

• GradSentry: Gradient Spectral Entropy for Backdoor Sample Filtering in Large Language Model Fine-Tuning
  Haodong Zhao, Tianyi Xu, Tianhang Zhao, Zhuosheng Zhang^†, Gongshen Liu^†
  Code
• ProtegoFed: Backdoor-Free Federated Instruction Tuning with Interspersed Poisoned Data
  Haodong Zhao, Jinming Hu, Zhaomin Wu^†, Zongru Wu, Wei Du, Junyi Hou, Caibei Zhao, Zhuosheng Zhang, Bingsheng He, Gongshen Liu^†
  Code
• EmbTracker: Traceable Black-box Watermarking for Federated Language Models
  Haodong Zhao, Jinming Hu, Yijie Bai, Tian Dong, Wei Du, Zhuosheng Zhang^†, Yanjiao Chen, Haojin Zhu, Gongshen Liu^†
• Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models
  Tianhang Zhao^*, Haodong Zhao^*, Wei Du^*, Pengzhou Cheng, Junxian Li, Sufeng Duan^†, Haojin Zhu, Gongshen Liu^†
  Code
• UniDial-EvalKit: A Unified Toolkit for Evaluating Multi-Faceted Conversational Abilities
  Qi Jia, Haodong Zhao, Dun Pei, Xiujie Song, Shibo Wang, Zijian Chen, Zicheng Zhang, Xiangyang Zhu, Guangtao Zhai
  Code
• Security of Large Model-based Agents: A Survey on Adversarial, Poisoning, and Backdoor Attacks
  Xinyun Liu, Zelei Cheng, Haodong Zhao, Ronghua Xu
• When Disagreements Elicit Robustness: Investigating Self-Repair Capabilities under LLM Multi-Agent Disagreements
  Tianjie Ju, Bowen Wang, Hao Fei, Mong-Li Lee, Wynne Hsu, Yun Li, Qianren Wang, Pengzhou Cheng, Zongru Wu, Haodong Zhao, Zhuosheng Zhang, Gongshen Liu
• Vertical Federated Learning in Practice: The Good, the Bad, and the Ugly
  Zhaomin Wu, Zhen Qin, Junyi Hou, Haodong Zhao, Qinbin Li, Bingsheng He, Lixin Fan
• TrojanRAG: Retrieval-Augmented Generation Can Be Backdoor Driver in Large Language Models
  Pengzhou Cheng, Yidong Ding, Tianjie Ju, Zongru Wu, Wei Du, Haodong Zhao, Haiyang Yu, Ping Yi, Zhuosheng Zhang, Gongshen Liu
  Code

2026

• IAG: Input-aware Backdoor Attack on VLM-based Visual Grounding
  Junxian Li^*, Beining Xu^*, Simin Chen, Jiatong Li, Jingdi Lei, Haodong Zhao^†, Di Zhang^†
  CVPR 2026
  Code
• Disagreements in Reasoning: How a Model's Thinking Process Dictates Persuasion in Multi-Agent Systems
  Haodong Zhao^*, Jidong Li^*, Zhaomin Wu^†, Tianjie Ju, Zhuosheng Zhang, Bingsheng He, Gongshen Liu^†
  LaMAS@AAAI'26 (Oral 🏆🏆🏆)
  Workshop Website
• Revisiting Backdoor Threat in Federated Instruction Tuning from a Signal Aggregation Perspective
  Haodong Zhao, Jinming Hu, Gongshen Liu
  ICASSP 2026
• Thinking in a Crowd: How Auxiliary Information Shapes LLM Reasoning
  Haodong Zhao^*, Chenyan Zhao^*, Yansi Li, Zhuosheng Zhang^†, Gongshen Liu^†
  ICASSP 2026
  Dataset
• LLM DNA: Tracing Model Evolution via Functional Representations
  Zhaomin Wu, Haodong Zhao, Ziyang Wang, Jizhou Guo, Qian Wang, Bingsheng He
  ICLR 2026 (Oral 🏆🏆🏆, top 1%)
  Website Code
• GhostEI-Bench: Do Mobile Agents Resilience to Environmental Injection in Dynamic On-Device Environments?
  Chiyu Chen, Xinhao Song, Yunkai Chai, Yang Yao, Haodong Zhao, Lijun Li, Jie Li, Yan Teng, Gongshen Liu, Yingchun Wang
  ICLR 2026
  Code
• OpenImplicit: Benchmarking Implicit Reasoning in MLLMs via Open-Ended Evaluation
  Jidong Li, Xiaofei Yin, Ya Guo, Shuheng Zhou, Yi Tu, Haodong Zhao, Sufeng Duan, Gongshen Liu, Huijia Zhu
  ICMR 2026
• Judge Before Answer: Can MLLM Discern the False Premise in Question?
  Jidong Li, Lingyong Fang, Haodong Zhao, Sufeng Duan, Gongshen Liu
  ICASSP 2026
  Code

2025

• FedRS-Bench: Realistic Federated Learning Datasets and Benchmarks in Remote Sensing
  Haodong Zhao^*, Peng Peng^*, Chiyu Chen, Linqing Huang^†, Gongshen Liu^†
  KDD 2025 workshop: FedKDD
  Website
• NSmark: Null Space Based Black-box Watermarking Defense Framework for Language Models
  Haodong Zhao, Jinming Hu, Peixuan Li, Fangqi Li, Jinrui Sha, Tianjie Ju, Peixuan Chen, Zhuosheng Zhang^†, Gongshen Liu^†
  ICLR 1st GenAI Watermarking Workshop, 2025
  Code
• Flooding Spread of Manipulated Knowledge in LLM-Based Multi-Agent Communities
  Tianjie Ju, Yiting Wang, Xinbei Ma, Pengzhou Cheng, Haodong Zhao, Yulong Wang, Lifeng Liu, Jian Xie, Zhuosheng Zhang, Gongshen Liu
  Science China Information Sciences (SCIS)
  Code
• Watch Out Your Album! On the Inadvertent Privacy Memorization in Multi-Modal Large Language Models
  Tianjie Ju, Yi Hua, Hao Fei, Zhenyu Shao, Yubin Zheng, Haodong Zhao, Mong-Li Lee, Wynne Hsu, Zhuosheng Zhang, Gongshen Liu
  ICML 2025
  Code
• Backdoor Attacks and Countermeasures in Natural Language Processing Models: A Comprehensive Security Review
  Pengzhou Cheng, Zongru Wu, Wei Du, Haodong Zhao, Wei Lu, Gongshen Liu
  TNNLS

2024

• Revisiting the Information Capacity of Neural Network Watermarks: Upper Bound Estimation and Beyond
  Fangqi Li, Haodong Zhao, Wei Du, Shilin Wang
  AAAI 2024
• UOR: Universal Backdoor Attacks on Pre-trained Language Models
  Wei Du, Peixuan Li, Haodong Zhao, Tianjie Ju, Ge Ren, Gongshen Liu
  ACL findings 2024
  Code
• NWS: NATURAL TEXTUAL BACKDOOR ATTACKS VIA WORD SUBSTITUTION
  Wei Du, Tongxin Yuan, Haodong Zhao, Gongshen Liu
  ICASSP 2024

2023

• FedPrompt: Communication-Efficient and Privacy Preserving Prompt Tuning in Federated Learning
  Haodong Zhao, Wei Du, Fangqi Li, Peixuan Li, Gongshen Liu
  ICASSP 2023
• PLMmark: A Secure and Robust Black-box Watermarking Framework for Pre-trained Language Models
  Peixuan Li, Pengzhou Cheng, Fangqi Li, Wei Du, Haodong Zhao, Gongshen Liu
  AAAI 2023
• Is Continuous Prompt a Combination of Discrete Prompts? Towards a Novel View for Interpreting Continuous Prompts
  Tianjie Ju, Yubin Zheng, Hanyi Wang, Haodong Zhao , Gongshen Liu
  ACL findings 2023
• A Universal Identity Backdoor Attack against Speaker Verification based on Siamese Network
  Haodong Zhao^*, Wei Du^*, Junjie Guo, Gongshen Liu^†
  Interspeech 2022

Talks

• [2026.03] Talk "Backdoor Attack and Defense in Federated Large Models" at 第四届中关村国际青年论坛 @ Zhongguancun Academy.
• [2026.03] Host Talk "拒绝“降智、减配、乱收费” - 面向 LLM API 的可信验证框架" at NICE. Speaker: Yanpei Guo. [video]
• [2026.02] Host Talk "给大模型做“亲子鉴定”：LLM 的 DNA 与模型演化图谱" at NICE. Speaker: Zhaomin Wu. [video]
• [2026.02] Host Talk "PaperDebugger：将 Agent 能力带入 Overleaf" at NICE. Speaker: Junyi Hou. [video]
• [2026.01] Talk "Disagreements in Reasoning: How a Model's Thinking Process Dictates Persuasion in Multi-Agent Systems" at WMAC 2026 @ AAAI. [slides]
• [2025.11] Host talk "ColorAgent：构建稳健、个性化、交互式的OS智能体" at NICE. Speaker: Ning Li. [video]
• [2023.09] Talk "Privacy computing in the era of large models" at Inclusion 2023. [slides]
• [2021.07] Host Talk "当代青年对话智能农业新未来" at WAIC 2021.

Mentorship

It is my honor to work with many talented students!

PhD Students

• Chiyu Chen (2025-2026), bachelor's thesis and co-authored 2 papers

Master Students

• Tianhang Zhao (2025-), co-authored 2 papers
• Jidong Li (2025-2026), co-authored 3 papers

Undergraduate Students

• Jinming Hu (2024-, incoming master student in our group), co-authored 6 papers
• Ziwu Liu (2025-), co-authored 1 patent
• Tianyi Xu (2025-), co-authored 1 paper
• Chenyan Zhao (2025), co-authored 1 paper
• Peng Peng (2025), co-authored 1 paper

Services

Professional Services

• Conference Area Chair (AC): ACL ARR 2026 March
• Journal Reviewer: TDSC, TIFS, TNNLS, TKDE, TKDD, TMM, SCIS, Neurocomputing
• Conference Reviewer: ICLR, ICML, NeurIPS, CVPR, AAAI, ARR, KDD, ACM MM, IJCAI, ICME, AISTATS, IJCNN
• Teaching Assistant for "Introduction to AI", Fall 2023; "Operating System" , Fall 2022, Spring 2023, Spring 2024; "Principles of Computer Viruses" , Spring 2022; "Computer Programming Practice" , Summer 2022, Summer 2023.

Community Services

• Chair of CCF Student Chapter of Shanghai Jiaotong University.
• Committe member of NICE, an NLP Academic Exchange Platform.
• Committe member of MLNLP, an NLP academic community.
• Part-time Counselor in SJTU.