Dong Xiaoning

PhD

Zhongguancun National Laboratory
Shichuang Yigu, Cuihu Sci-Tech Park
Haidian District, Beijing 100194, P.R. China

Lab: Building 13
Email: dongxn02 at gmail dot com

Biography

I am currently an assistant researcher at Zhongguancun Laboratory (ZGC Lab). I obtained my PhD degree from Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University, advised by Prof. Xu Wei. Previously, I obtained my master's degree in computer science from National University of Defense Technology and my bachelor's degree from Xidian University.

My research focuses on the safety and reliability of large language models (LLMs), including jailbreak attacks, multi-agent security, multimodal safety, and factuality alignment. I am also interested in LLM-based code generation, particularly for privacy-preserving computation.

🔥 News

[08/2026] One paper accepted at ACM TOSEM: SPDZCoder: Combining Expert Knowledge with LLMs for Generating Privacy-Computing Code.
[06/2026] One paper accepted at ACL Findings 2026: FAITH: Factuality Alignment through Integrating Trustworthiness and Honestness.
[05/2026] I was awarded the IJCNN 2026 Travel Grant by the International Neural Network Society (INNS).
[04/2026] One paper accepted at IJCNN 2026: ReaSafe: Reasoning-Centric Safety Alignment for Multimodal LLMs.
[01/2026] Serving as Conference Reviewer for NeurIPS 2026.
[09/2025] One paper accepted at EMNLP 2025: SURE: Safety Understanding and Reasoning Enhancement for Multimodal Large Language Models.
[05/2025] One paper accepted at ACL 2025: SATA: A Paradigm for LLM Jailbreak via Simple Assistive Task Linkage.

Publications [show selected / show all by date]

* co-first author ✉ corresponding author

Industry Experience

Beijing Institute of Control and Electronic Technology, CASIC
System Engineer, Jul. 2017 – Mar. 2020
Contributed to defense equipment R&D projects, including ship target recognition in SAR/infrared imagery and telemetry ground station software systems.

Tencent Technology (Beijing) Co., Ltd.
Project Manager, Apr. 2020 – Aug. 2020
Managed university research collaboration projects; worked with research and technical teams to establish and maintain academic partnerships.

Internship

TusStar Venture Capital
Investment Assistant Intern, Jun. 2026 – Jul. 2026
Sourced early-stage deal flow and connected with high-quality startups; assisted with project screening, due diligence working papers, and meeting minutes.

Amazon (China) Holding Company Limited
Applied Scientist Intern, May 2023 – Nov. 2023
Researched instruction data augmentation for LLMs; built a pipeline for Amazon product search and customer service dialogues; results integrated into LLM fine-tuning pipeline for Amazon Japan.

Microsoft Research Asia
Research Intern, Aug. 2022 – Dec. 2022
Researched reinforcement learning and multi-agent RL; participated in NeurIPS 2022 CityLearn Challenge on AI for energy management, achieving 5th place (top 5/100+).

Noah's Ark Lab, Huawei Central Research Institute
Research Intern, Jul. 2021 – Dec. 2021
Proposed a non-autoregressive machine translation method based on cross-lingual pre-trained models (XLM) with iterative decoding.

Selected Prizes & Awards

National Scholarship (top 1%)
Top Prize Scholarship of Xidian University (top 1%)
National Encouragement Scholarship
Huawei Scholarship; China Telecom Scholarship
Social Work Scholarship, IIIS, Tsinghua University (×2)
NUDT Graduate Scholarship (×2); First Prize Scholarship, Xidian University (×2); Outstanding Graduate, Xidian University
NeurIPS 2022 CityLearn Challenge, 5th/100+
Outstanding CCP Branch Secretary & Member, Tsinghua University
China Computer Federation Outstanding Undergraduate Award (CCF OUA)

Academic Service

NeurIPS 2026 Conference Reviewer
ACL 2026 Workshop C3NLP Reviewer
IJCNN 2026 Reviewer
ACM TOSEM Reviewer