Haokun Chen

About Me

I received my Ph.D. from LMU Munich in Germany, advised by Prof. Dr. Volker Tresp and co-supervised by Dr. Denis Krompass (Siemens AG). During my doctoral studies, I completed internships at Amazon and Intel. Besides, I earned my Master’s degree from TUM in Germany and my Bachelor’s degree from Tongji University in China. I am trilingual in Mandarin, English, and German, and I enjoy playing basketball, swimming, and hip-hop dancing.

Research Interests

Generative AI: Post-training of Large Language Models (LLMs), Vision-Language Models.
Distritbuted Model Training: Data heterogeneity & deficiency in Federated Learning.
Trustworthy AI: Adversarial Attacks, Machine Unlearning.

Education

Nov 2021 - Nov 2025 Ph.D. Computer Science, Ludwig Maximilian University of Munich, Germany.
Oct 2018 - Nov 2021 M.Sc Computer Science, Technical University of Munich, Germany.
Oct 2014 - Oct 2018 B.Sc Mechatronic Engineering, Tongji University, China.

Experience

Jun 2025 - Dec 2025 Applied Scientist Intern, Amazon, Berlin, Germany.
Aug 2024 - Nov 2024 Research Scientist Intern, Intel, Remote, USA.
Nov 2021 - Jun 2025 Doctorate Researcher, Siemens AG, Munich, Germany.
Mar 2020 - Nov 2020 Research Intern, BMW Autonomous Driving Campus, Munich, Germany.

Publications

AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language Models

Haokun Chen, Jianing Li, Yao Zhang, Jinhe Bi, Yan Xia, Jindong Gu, Volker Tresp

The 40th Annual AAAI Conference on Artificial Intelligence (AAAI), Jan. 2026.

Code PDF Poster Presentation
Soft Token Attacks Cannot Reliably Audit Unlearning in Large Language Models

Haokun Chen, Sebastian Szyller, Weilin Xu, Nageen Himayat

The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov. 2025.

PDF Poster Presentation
FedBiP: Heterogeneous One-Shot Federated Learning with Personalized Latent Diffusion Models

Haokun Chen, Hang Li, Yao Zhang, Gengyuan Zhang, Jinhe Bi, Philip Torr, Jindong Gu, Yueqi Zhang, Denis Krompass, Volker Tresp

The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2025.

Code PDF Poster Presentation
FedPop: Federated Population-based Hyperparameter Tuning

Haokun Chen, Denis Krompass, Jindong Gu, Volker Tresp

The 39th Annual AAAI Conference on Artificial Intelligence (AAAI), Feb. 2025.

Code PDF Oral Presentation
FedDAT: An Approach for Foundation Model Finetuning in Multi-Modal Heterogeneous Federated Learning

Haokun Chen, Yao Zhang, Denis Krompass, Jindong Gu, Volker Tresp

The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), Feb. 2024.

Code PDF Poster Presentation
FRAug: Tackling Federated Learning with Non-IID Features via Representation Augmentation

Haokun Chen, Ahmed Frikha, Denis Krompass, Jindong Gu, Volker Tresp

International Conference on Computer Vision (ICCV), Oct. 2023.

Code PDF Poster Presentation
Towards Data-free Domain Generalization

Ahmed Frikha ^* , Haokun Chen, Denis Krompass, Volker Tresp, Thomas Runkler

14th Asian Conference on Machine Learning (ACML), Dec. 2022.

Code PDF Poster Presentation
LLaVA Steering: Visual Instruction Tuning with 500x Fewer Parameters through Modality Linear Representation-Steering

Jinhe Bi, Yujun Wang, Haokun Chen, Xun Xiao, Artur Hecker, Volker Tresp, Yunpu Ma,

The 63rd Annual Meeting of the Association for Computational Linguistics (ACL), Jul. 2025

PDF Poster Presentation
CL-CrossVQA: A Continual Learning Benchmark for Cross-Domain Visual Question Answering

Yao Zhang, Haokun Chen, Ahmed Frikha, Yezi Yang, Denis Krompass, Gengyuan Zhang, Jindong Gu, Volker Tresp

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Feb. 2025

PDF Poster Presentation