← Search

Jingxiao Yang

2 accepted papers

2026

HogVul: Black-box Adversarial Code Generation Framework Against LM-based Vulnerability Detectors

AAAI 2026technical

Recent advances in software vulnerability detection have been driven by Language Model (LM)-based approaches. However, these models remain vulnerable to adversarial attacks that exploit lexical and syntax perturbations, allowing critical flaws to evade detection. Existing black-box attacks on LM-bas

Cited by 0SourcePDFScholar
2026

Mitigating Manifold Departure: Uncertainty-aware Subspace Rectification for Trustworthy MLLM Decoding

ICML 2026poster

Multimodal Large Language Models often suffer from object hallucinations, where generated outputs are inconsistent with the visual evidence. This issue is typically attributed to the over-reliance on language priors, which can override the visual context. Recent training-free decoding strategies add…

Cited by 0SourceScholar