2 papers
cs.CR2026
Transferable Backdoor Attacks for Code Models via Sharpness-Aware Adversarial Perturbation
Shuyu Chang, Haiping Huang, Yanjun Zhang +3
Code models are increasingly adopted in software development but remain vulnerable to backdoor attacks via poisoned training data. Existing backdoor attacks on code models face a f…
cs.CL2025
A Large Language Model Guided Topic Refinement Mechanism for Short Text Modeling
Shuyu Chang, Rui Wang, Peng Ren +2
Modeling topics effectively in short texts, such as tweets and news snippets, is crucial to capturing rapidly evolving social trends. Existing topic models often struggle to accura…