Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Yue Duan

Yue Duan

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

Crypto×2AI×1ML×1Software Eng.×1

Frequent co-authors

Hyo Seo Kim1×
Gang Luo1×
Can Chen1×
Binghui Wang1×
Ren Wang1×
Reinelle Jan Bugnot1×

Research Timeline

2026
Compositional Jailbreaking: An Empirical Analysis of Mutator Chain Interactions in Aligned LLMs

This paper systematically analyzes the interaction of multiple weak jailbreak attacks (mutators) applied sequentially to LLMs, finding that most combinations fail due to destructive interference, revealing structural properties of model safety alignment.

MoCo-EA: Exploiting Adversarial Mode Connectivity for Efficient Evolutionary Attacks

The paper introduces MoCo-EA, an evolutionary attack method that replaces standard crossover with a continuous Bézier curve interpolation to efficiently exploit the connected manifold structure of adversarial examples.

Highlighted terms show continued research focus across papers

Papers

cs.CRcs.AIcs.LGRecentMay 18, 2026

MoCo-EA: Exploiting Adversarial Mode Connectivity for Efficient Evolutionary Attacks

Hyo Seo Kim, Gang Luo, Can Chen, Binghui Wang +2 more

The paper introduces MoCo-EA, an evolutionary attack method that replaces standard crossover with a continuous Bézier curve interpolation to efficiently exploit the connected manifold structure of adv…

View →
cs.CRcs.SERecentMay 15, 2026

Compositional Jailbreaking: An Empirical Analysis of Mutator Chain Interactions in Aligned LLMs

Reinelle Jan Bugnot, Soohyeon Choi, Hoon Wei Lim, Yue Duan

This paper systematically analyzes the interaction of multiple weak jailbreak attacks (mutators) applied sequentially to LLMs, finding that most combinations fail due to destructive interference, reve…

View →