Profiles2022-2025
Xiangyu Qi
Princeton University
7Papers
0Last author
11Unique co-authors
5First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers7
- Defensive Prompt Patch: A Robust and Generalizable Defense of Large Language Models against Jailbreak AttacksAuthor 2 of 4ACL Findings 2025 · 9 citations
- On Evaluating the Durability of Safeguards for Open-Weight LLMsAuthor 1 of 10ICLR 2025
- Safety Alignment Should be Made More Than Just a Few Tokens DeepAuthor 1 of 7ICLR 2025
- Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!Author 1 of 7ICLR 2024 · 1,104 citations
- BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input DetectionAuthor 2 of 6ICLR 2024 · 20 citations
- Revisiting the Assumption of Latent Separability for Backdoor DefensesAuthor 1 of 5ICLR 2023
- Towards Practical Deployment-Stage Backdoor Attack on Deep Neural NetworksAuthor 1 of 6CVPR 2022
