ResearchAlpha Leak

Conferences Topics Top Authors Rankings Browse All

Home/Authors/Xiangyu Qi

Xiangyu Qi

Topic trends: 32,543 papers · similarity ≥ 0.4 · year ≥ 2024 · Data sourced from Semantic Scholar

34,598 papers | Abstracts: 31,650 (91.5%) | Citations: 34,598 (100.0%) | arXiv: 26,074 (75.4%)

Built: Feb 14, 2026, 11:22 PM AMS

6

papers

1,693

total citations

papers (6)

Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Safety Alignment Should be Made More Than Just a Few Tokens Deep

Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications

SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal

Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networks

BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection