Publications | Zhaomin Wu

Selected Publications

Phylogenetic tree of LLM families reconstructed from functional DNA distances.

LLM DNA: Tracing Model Evolution via Functional Representations

Zhaomin Wu, Haodong Zhao, Ziyang Wang, Jizhou Guo, Qian Wang, Bingsheng He

ICLR 2026

Oral Presentation (1%)

A foundation for understanding model silos: compact functional representations reveal relationships among heterogeneous and black-box language models.

Paper Code Website

@inproceedings{wu2026llmdna,
  author = {Wu, Zhaomin and Zhao, Haodong and Wang, Ziyang and Guo, Jizhou and Wang, Qian and He, Bingsheng},
  title = {LLM DNA: Tracing Model Evolution via Functional Representations},
  booktitle = {The Fourteenth International Conference on Learning Representations (ICLR)},
  year = {2026},
  url = {https://openreview.net/forum?id=UIxHaAqFqQ},
}

Distribution of LLMs across Truthfulness, Guessing, Hallucination, and Deception regions based on deceptive intention and behavior scores.

Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts

Zhaomin Wu, Mingzhe Du, See-Kiong Ng, Bingsheng He

ICLR 2026

Oral Presentation (1%)

Extends my work on model understanding to safety, examining when language models express beliefs inconsistently even under benign prompts.

Paper

@inproceedings{wu2026deception,
  author = {Wu, Zhaomin and Du, Mingzhe and Ng, See-Kiong and He, Bingsheng},
  title = {Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts},
  booktitle = {The Fourteenth International Conference on Learning Representations (ICLR)},
  year = {2026},
  url = {https://openreview.net/forum?id=PDBBYwd1LY},
}

DeltaBoost removes deleted records from a trained model, compared with retraining after database deletion.

DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning

Zhaomin Wu, Junhui Zhu, Qinbin Li, Bingsheng He

SIGMOD 2023

Honorable Mention for Best Artifact (Top 3)

Connects machine unlearning with practical systems, enabling efficient data deletion from gradient-boosted decision trees.

Paper Code

@article{wu2023deltaboost,
  author = {Wu, Zhaomin and Zhu, Junhui and Li, Qinbin and He, Bingsheng},
  title = {DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning},
  year = {2023},
  issue_date = {June 2023},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {1},
  number = {2},
  url = {https://doi.org/10.1145/3589313},
  doi = {10.1145/3589313},
  journal = {Proc. ACM Manag. Data},
  articleno = {168},
  numpages = {26},
  keywords = {data deletion, gradient boosting decision trees, machine unlearning},
}

VertiBench expands the range of party importance balance and inter-party correlation covered by VFL benchmarks.

VertiBench: Advancing Feature Distribution Diversity in Vertical Federated Learning Benchmarks

Zhaomin Wu, Junyi Hou, Bingsheng He

ICLR 2024

Builds research infrastructure for practical vertical federated learning by making diverse feature distributions measurable and reproducible.

Paper Code Website

@inproceedings{wu2024vertibench,
  author = {Wu, Zhaomin and Hou, Junyi and He, Bingsheng},
  title = {VertiBench: Advancing Feature Distribution Diversity in Vertical Federated Learning Benchmarks},
  booktitle = {The Twelfth International Conference on Learning Representations},
  year = {2024},
  url = {https://openreview.net/forum?id=glwwbaeKm2},
}

Neural network architecture of FedSim: local and aggregate models, a similarity weight gate, a sort gate, and a CNN merge model.

A Coupled Design of Exploiting Record Similarity for Practical Vertical Federated Learning

Zhaomin Wu, Qinbin Li, Bingsheng He

NeurIPS 2022

Establishes my research on practical data silos by coupling record linkage with federated training when exact identifiers are unavailable.

Paper Code

@inproceedings{wu2022coupledvfl,
  author = {Wu, Zhaomin and Li, Qinbin and He, Bingsheng},
  editor = {Koyejo, S. and Mohamed, S. and Agarwal, A. and Belgrave, D. and Cho, K. and Oh, A.},
  booktitle = {Advances in Neural Information Processing Systems},
  pages = {21087--21100},
  publisher = {Curran Associates, Inc.},
  title = {A Coupled Design of Exploiting Record Similarity for Practical Vertical Federated Learning},
  url = {https://proceedings.neurips.cc/paper\_files/paper/2022/file/84b744165a0597360caad96b06e69313-Paper-Conference.pdf},
  volume = {35},
  year = {2022},
}

All Publications

* Equal contribution · † Corresponding author

Peer-reviewed Publications

2026

Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts

Zhaomin Wu, Mingzhe Du, See-Kiong Ng, Bingsheng He

ICLR 2026 Paper

Oral Presentation (1%)

@inproceedings{wu2026deception,
  author = {Wu, Zhaomin and Du, Mingzhe and Ng, See-Kiong and He, Bingsheng},
  title = {Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts},
  booktitle = {The Fourteenth International Conference on Learning Representations (ICLR)},
  year = {2026},
  url = {https://openreview.net/forum?id=PDBBYwd1LY},
}

EmoTrack: Clinical-Semantic Modeling for Text-Based Depression Severity Estimation

Zhaomin Wu, Jiayi Li, Bingsheng He

NeurIPS 2026 Paper

@inproceedings{wu2026emotrack,
  author = {Wu, Zhaomin and Li, Jiayi and He, Bingsheng},
  title = {EmoTrack: Clinical-Semantic Modeling for Text-Based Depression Severity Estimation},
  booktitle = {Advances in Neural Information Processing Systems},
  year = {2026},
  url = {https://arxiv.org/abs/2605.22286},
}

LLM DNA: Tracing Model Evolution via Functional Representations

Zhaomin Wu, Haodong Zhao, Ziyang Wang, Jizhou Guo, Qian Wang, Bingsheng He

ICLR 2026 Paper Code Website

Oral Presentation (1%)

@inproceedings{wu2026llmdna,
  author = {Wu, Zhaomin and Zhao, Haodong and Wang, Ziyang and Guo, Jizhou and Wang, Qian and He, Bingsheng},
  title = {LLM DNA: Tracing Model Evolution via Functional Representations},
  booktitle = {The Fourteenth International Conference on Learning Representations (ICLR)},
  year = {2026},
  url = {https://openreview.net/forum?id=UIxHaAqFqQ},
}

WikiDBGraph: A Data Management Benchmark Suite for Collaborative Learning over Database Silos

Zhaomin Wu^*, Ziyang Wang^*, Bingsheng He

ICDE 2026 Paper

@inproceedings{wu2026wikidbgraph,
  author = {Wu, Zhaomin and Wang, Ziyang and He, Bingsheng},
  title = {WikiDBGraph: A Data Management Benchmark Suite for Collaborative Learning over Database Silos},
  booktitle = {Proceedings of the 42th International Conference on Data Engineering},
  year = {2026},
}

Mining Intrinsic Rewards from LLM Hidden States for Efficient Best-of-N Sampling

Jizhou Guo, Zhaomin Wu^†, Hanchen Yang, Philip S. Yu

KDD 2026 Paper

@inproceedings{guo2026intrinsicrewards,
  author = {Guo, Jizhou and Wu, Zhaomin and Yang, Hanchen and Yu, Philip S.},
  title = {Mining Intrinsic Rewards from LLM Hidden States for Efficient Best-of-N Sampling},
  booktitle = {Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)},
  year = {2026},
  url = {https://dl.acm.org/doi/10.1145/3770854.3780302},
}

Reasoning or Rambling? Exploring the Effect of Thinking on Agent Persuasion

Haodong Zhao, Jidong Li, Zhaomin Wu^†, Tianjie Ju, Zhuosheng Zhang, Bingsheng He, Gongshen Liu^†

EMNLP 2026 Findings Paper

@inproceedings{zhao2026reasoning,
  author = {Zhao, Haodong and Li, Jidong and Wu, Zhaomin and Ju, Tianjie and Zhang, Zhuosheng and He, Bingsheng and Liu, Gongshen},
  title = {Reasoning or Rambling? Exploring the Effect of Thinking on Agent Persuasion},
  booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2026},
  year = {2026},
  url = {https://openreview.net/forum?id=g7pOyKiWO5},
}

CrossAlpha: An Annual-Report Benchmark for Cross-Market Factor Research

Qian Wang, Zhongyi Tong, Nuo Chen, Zhaomin Wu, Bingsheng He

EMNLP 2026 Findings Paper

@inproceedings{wang2026crossalpha,
  author = {Wang, Qian and Tong, Zhongyi and Chen, Nuo and Wu, Zhaomin and He, Bingsheng},
  title = {CrossAlpha: An Annual-Report Benchmark for Cross-Market Factor Research},
  booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2026},
  year = {2026},
  url = {https://openreview.net/forum?id=nNZ2BEfGUb},
}

Personalized Federated Fine-Tuning for LLMs via Data-Driven Heterogeneous Model Architectures

Yicheng Zhang, Zhen Qin, Zhaomin Wu, Shuiguang Deng^†

WWW 2026 Paper

Oral Presentation (9%)

@inproceedings{zhang2026personalizedfed,
  author = {Zhang, Yicheng and Qin, Zhen and Wu, Zhaomin and Deng, Shuiguang},
  title = {Personalized Federated Fine-Tuning for LLMs via Data-Driven Heterogeneous Model Architectures},
  booktitle = {Proceedings of the ACM on Web Conference 2026},
  year = {2026},
  url = {https://dl.acm.org/doi/10.1145/3774904.3792147},
}

2025

Model-based Large Language Model Customization as Service

Zhaomin Wu^*, Jizhou Guo^*, Junyi Hou, Bingsheng He, Lixin Fan, Qiang Yang

EMNLP 2025 Paper

@inproceedings{wu2025llmcustomization,
  author = {Wu, Zhaomin and Guo, Jizhou and Hou, Junyi and He, Bingsheng and Fan, Lixin and Yang, Qiang},
  title = {Model-based Large Language Model Customization as Service},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year = {2025},
  url = {https://aclanthology.org/2025.emnlp-main.248.pdf},
}

Federated Data-Efficient Instruction Tuning for Large Language Models

Zhen Qin, Zhaomin Wu^†, Bingsheng He, Shuiguang Deng^†

ACL 2025 Findings Paper

@inproceedings{qin2025instructiontuning,
  author = {Qin, Zhen and Wu, Zhaomin and He, Bingsheng and Deng, Shuiguang},
  title = {Federated Data-Efficient Instruction Tuning for Large Language Models},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2025},
  year = {2025},
  url = {https://aclanthology.org/2025.findings-acl.803/},
}

2024

Federated Transformer: Multi-Party Vertical Federated Learning on Practical Fuzzily Linked Data

Zhaomin Wu, Junyi Hou, Yiqun Diao, Bingsheng He

NeurIPS 2024 Paper Code

@inproceedings{wu2024federatedtransformer,
  author = {Wu, Zhaomin and Hou, Junyi and Diao, Yiqun and He, Bingsheng},
  booktitle = {Advances in Neural Information Processing Systems},
  publisher = {Curran Associates, Inc.},
  title = {Federated Transformer: Multi-Party Vertical Federated Learning on Practical Fuzzily Linked Data},
  volume = {36},
  year = {2024},
  url = {https://openreview.net/forum?id=FqWyzyErVT},
}

VertiBench: Advancing Feature Distribution Diversity in Vertical Federated Learning Benchmarks

Zhaomin Wu, Junyi Hou, Bingsheng He

ICLR 2024 Paper Code Website

@inproceedings{wu2024vertibench,
  author = {Wu, Zhaomin and Hou, Junyi and He, Bingsheng},
  title = {VertiBench: Advancing Feature Distribution Diversity in Vertical Federated Learning Benchmarks},
  booktitle = {The Twelfth International Conference on Learning Representations},
  year = {2024},
  url = {https://openreview.net/forum?id=glwwbaeKm2},
}

2023

DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning

Zhaomin Wu, Junhui Zhu, Qinbin Li, Bingsheng He

SIGMOD 2023 Paper Code

Honorable Mention for Best Artifact (Top 3)

@article{wu2023deltaboost,
  author = {Wu, Zhaomin and Zhu, Junhui and Li, Qinbin and He, Bingsheng},
  title = {DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning},
  year = {2023},
  issue_date = {June 2023},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {1},
  number = {2},
  url = {https://doi.org/10.1145/3589313},
  doi = {10.1145/3589313},
  journal = {Proc. ACM Manag. Data},
  articleno = {168},
  numpages = {26},
  keywords = {data deletion, gradient boosting decision trees, machine unlearning},
}

FedTree: A Federated Learning System for Trees

Qinbin Li, Zhaomin Wu, Yanzheng Cai, Yuxuan Han, Ching Man Yung, Tianyuan Fu, Bingsheng He

MLSys 2023 Paper Code

@inproceedings{li2023fedtree,
  author = {Li, Qinbin and Wu, Zhaomin and Cai, Yanzheng and Han, Yuxuan and Yung, Ching Man and Fu, Tianyuan and He, Bingsheng},
  editor = {Song, D. and Carbin, M. and Chen, T.},
  booktitle = {Proceedings of Machine Learning and Systems},
  pages = {89--103},
  publisher = {Curan},
  title = {FedTree: A Federated Learning System for Trees},
  url = {https://proceedings.mlsys.org/paper\_files/paper/2023/file/3430e7055936cb8e26451ed49fce84a6-Paper-mlsys2023.pdf},
  volume = {5},
  year = {2023},
}

2022

A Coupled Design of Exploiting Record Similarity for Practical Vertical Federated Learning

Zhaomin Wu, Qinbin Li, Bingsheng He

NeurIPS 2022 Paper Code

@inproceedings{wu2022coupledvfl,
  author = {Wu, Zhaomin and Li, Qinbin and He, Bingsheng},
  editor = {Koyejo, S. and Mohamed, S. and Agarwal, A. and Belgrave, D. and Cho, K. and Oh, A.},
  booktitle = {Advances in Neural Information Processing Systems},
  pages = {21087--21100},
  publisher = {Curran Associates, Inc.},
  title = {A Coupled Design of Exploiting Record Similarity for Practical Vertical Federated Learning},
  url = {https://proceedings.neurips.cc/paper\_files/paper/2022/file/84b744165a0597360caad96b06e69313-Paper-Conference.pdf},
  volume = {35},
  year = {2022},
}

Practical Vertical Federated Learning with Unsupervised Representation Learning

Zhaomin Wu, Qinbin Li, Bingsheng He

TBD 2022 Paper Code

@article{wu2022fedonce,
  author = {Wu, Zhaomin and Li, Qinbin and He, Bingsheng},
  journal = {IEEE Transactions on Big Data},
  title = {Practical Vertical Federated Learning with Unsupervised Representation Learning},
  year = {2022},
  volume = {},
  number = {01},
  issn = {2332-7790},
  pages = {1-1},
  keywords = {collaborative work;privacy;differential privacy;training;data privacy;costs;unsupervised learning},
  doi = {10.1109/TBDATA.2022.3180117},
  publisher = {IEEE Computer Society},
  address = {Los Alamitos, CA, USA},
  url = {https://doi.org/10.1109/TBDATA.2022.3180117},
}

The OARF Benchmark Suite: Characterization and Implications for Federated Learning Systems

Sixu Hu, Yuan Li, Xu Liu, Qinbin Li, Zhaomin Wu, Bingsheng He

TIST 2022 Paper Code

@article{hu2022oarf,
  author = {Hu, Sixu and Li, Yuan and Liu, Xu and Li, Qinbin and Wu, Zhaomin and He, Bingsheng},
  title = {The OARF Benchmark Suite: Characterization and Implications for Federated Learning Systems},
  year = {2022},
  issue_date = {August 2022},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {13},
  number = {4},
  issn = {2157-6904},
  url = {https://doi.org/10.1145/3510540},
  doi = {10.1145/3510540},
  journal = {ACM Trans. Intell. Syst. Technol.},
  articleno = {63},
  numpages = {32},
  keywords = {framework, dataset, benchmark, machine learning, Federated learning},
}

A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection

Qinbin Li, Zeyi Wen, Zhaomin Wu, Sixu Hu, Naibo Wang, Yuan Li, Xu Liu, Bingsheng He

TKDE 2022 Paper

@article{li2022flsurvey,
  author = {Li, Qinbin and Wen, Zeyi and Wu, Zhaomin and Hu, Sixu and Wang, Naibo and Li, Yuan and Liu, Xu and He, Bingsheng},
  journal = {IEEE Transactions on Knowledge \&amp; Data Engineering},
  title = {A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection},
  year = {2022},
  volume = {35},
  number = {04},
  issn = {1558-2191},
  pages = {3347-3366},
  keywords = {collaborative work;data models;machine learning;data privacy;computational modeling;deep learning;servers},
  doi = {10.1109/TKDE.2021.3124599},
  publisher = {IEEE Computer Society},
  address = {Los Alamitos, CA, USA},
  url = {https://doi.org/10.1109/TKDE.2021.3124599},
}

2020

Privacy-Preserving Gradient Boosting Decision Trees

Qinbin Li, Zhaomin Wu, Zeyi Wen, Bingsheng He

AAAI 2020 Paper Code

@inproceedings{li2020ppgbdt,
  author = {Li, Qinbin and Wu, Zhaomin and Wen, Zeyi and He, Bingsheng},
  title = {Privacy-Preserving Gradient Boosting Decision Trees},
  booktitle = {The Thirty-Fourth {AAAI} Conference on Artificial Intelligence},
  pages = {784--791},
  publisher = {{AAAI} Press},
  year = {2020},
  url = {https://doi.org/10.1609/aaai.v34i01.5422},
  doi = {10.1609/AAAI.V34I01.5422},
  timestamp = {Sat, 30 Sep 2023 09:33:11 +0200},
  biburl = {https://dblp.org/rec/conf/aaai/LiWWH20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org},
}

Preprints & Ongoing Work

2026

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^†

arXiv 2026 Paper Code

@article{zhao2026protegofed,
  author = {Zhao, Haodong and Hu, Jinming and Wu, Zhaomin and Wu, Zongru and Du, Wei and Hou, Junyi and Zhao, Caibei and Zhang, Zhuosheng and He, Bingsheng and Liu, Gongshen},
  title = {ProtegoFed: Backdoor-Free Federated Instruction Tuning with Interspersed Poisoned Data},
  journal = {arXiv preprint arXiv:2603.00516},
  year = {2026},
  url = {https://arxiv.org/abs/2603.00516},
}

LongCounsel-8: A Benchmark Suite for Longitudinal Depression Tracking from Multi-Session Counseling Dialogues

Jiayi Li, Zhaomin Wu, Bingsheng He

arXiv 2026 Paper

@article{li2026longcounsel8,
  author = {Li, Jiayi and Wu, Zhaomin and He, Bingsheng},
  title = {LongCounsel-8: A Benchmark Suite for Longitudinal Depression Tracking from Multi-Session Counseling Dialogues},
  journal = {arXiv preprint arXiv:2609.03507},
  year = {2026},
  url = {https://arxiv.org/abs/2609.03507}
}

2025

Learning Relational Tabular Data without Shared Features

Zhaomin Wu, Shida Wang, Ziyang Wang, Bingsheng He

arXiv 2025 Paper

@article{wu2025relationaltabular,
  author = {Wu, Zhaomin and Wang, Shida and Wang, Ziyang and He, Bingsheng},
  title = {Learning Relational Tabular Data without Shared Features},
  journal = {arXiv preprint arXiv:2502.10125},
  year = {2025},
}

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

arXiv 2025 Paper

@article{wu2025vflpractice,
  author = {Wu, Zhaomin and Qin, Zhen and Hou, Junyi and Zhao, Haodong and Li, Qinbin and He, Bingsheng and Fan, Lixin},
  title = {Vertical Federated Learning in Practice: The Good, the Bad, and the Ugly},
  journal = {arXiv preprint arXiv:2502.08160},
  year = {2025},
}