Profiles2022-2026
Zhexin Zhang
The Conversational AI (CoAI ) group, DCST, Tsinghua University
7Papers
0Last author
4Unique co-authors
6First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers7
- How Should We Enhance the Safety of Large Reasoning Models: An Empirical StudyAuthor 1 of 13ACL 2026 · 20 citations
- Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!Author 1 of 7ICLR 2026 · 5 citations
- Knowledge-to-Jailbreak: Investigating Knowledge-driven Jailbreaking Attacks for Large Language ModelsAuthor 4 of 9KDD 2025
- ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety DetectorsAuthor 1 of 10EMNLP Findings 2024 · 14 citations
- Defending Large Language Models Against Jailbreaking Attacks Through Goal PrioritizationAuthor 1 of 6ACL 2024
- SafetyBench: Evaluating the Safety of Large Language ModelsAuthor 1 of 10ACL 2024
- Constructing Highly Inductive Contexts for Dialogue Safety through Controllable Reverse GenerationAuthor 1 of 8EMNLP Findings 2022 · 3 citations
