JailbreakSkill: Scaling Automated Red-Teaming with Reusable and Ever-Evolving Skills

Xiaoyu Wen*, Jiajia Li*, Zhida He, Peng Yu, Chenxu Wang, Han Qi, Ziyuan Zhou, Cheng Jin, Ying Wen, Xingcheng Xu, Shuyue Hu, Tianhang Zheng, Chaochao Lu, Qiaosheng Zhang†

arxiv preprint, 2026

[Paper] [Code]

Not All Turns Matter: Credit Assignment for Multi-Turn Jailbreaking

Zhida He*, Xiaoyu Wen*, Han Qi, Ziyuan Zhou, Peng Yu, Xingcheng Xu, Dongrui Liu, Xia Hu, Chaochao Lu, Qiaosheng Zhang†

arxiv preprint, 2026

[Paper] [Code]

MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety

Xiaoyu Wen*, Zhida He*, Han Qi, Ziyu Wan, Zhongtian Ma, Ying Wen, Tianhang Zheng, Xingcheng Xu, Chaochao Lu and Qiaosheng Zhang†

International Conference on Machine Learning (ICML), 2026

[Paper] [Code] CCF-A

Disentangling Intent from Role: Adversarial Self-Play for Persona-Invariant Safety Alignment

Jiajia Li*, Xiaoyu Wen*, Shuyue Hu, Qiaosheng Zhang† and Zhen Wang†;

International Conference on Machine Learning (ICML), 2026

[Paper] [Code] CCF-A

VAST: Video Ability-Stratified Taxonomy for Data-Efficient Video Reasoning

Zhongan Wang*, Xiaoyu Wen*, Lingxiao Du, Kun Li, zhiliang wu, Xingcheng Xu, Qiaosheng Zhang†, Chaochao Lu, Hehe Fan†

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

[Paper] [Page] CCF-A

ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learning

Ziyu Wan*; Yunxiang Li*; Xiaoyu Wen*; Yan Song; Hanjing Wang; Linyi Yang; Mark Schmidt; Jun Wang; Weinan Zhang; Shuyue Hu†; Ying Wen†

Neural Information Processing Systems (NeurIPS), 2025

[Paper] [Code] CCF-A

Contrastive Representation for Data Filtering in Cross-Domain Offline Reinforcement Learning

Xiaoyu Wen, Chenjia Bai†, Kang Xu, Xudong Yu, Yang Zhang, Xuelong Li, Zhen Wang†

International Conference on Machine Learning (ICML), 2024

[Paper] [Code] CCF-A

Towards Robust Offline-to-Online Reinforcement Learning via Uncertainty and Smoothness

Xiaoyu Wen*, Xudong Yu*, Rui Yang, Chenjia Bai†, Zhen Wang†

AAAI-24 Journal Track; Journal of Artificial Intelligence Research (JAIR) 81, 481-509, 2024

[Paper] [Code] CCF-B

SafeWork-R1: Co-evolving Safety and Intelligence under the AI-45◦ Law

Shanghai Artificial Intelligence Laboratory

arxiv preprint, 2025

An Early Warning of Emerging Biosecurity Risks in Frontier LLMs

Shanghai Artificial Intelligence Laboratory

arxiv preprint, 2026

[Paper]

Temporal Consistent Multi-View Perception for Robust Embodied Manipulation

Haoyuan Chen, Rushuai Yang, Junjie Zhang, Xiaoyu Wen, Yi Chen, Dengxiu Yu, Chenjia Bai†, Zhen Wang†

Pattern Recognition (PR), 112177, 2025

[Paper] CCF-B CAS Q1

Fair ASR: Re-Evaluating Black-Box Jailbreaks under Shared Target-Call Budgets

Zhida He, Xiaoyu Wen, Han Qi, Ziyuan Zhou, Peng Yu, Jiajia Li, Chaochao Lu, Qiaosheng Zhang†

arxiv preprint, 2026

[Paper] [Code]