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

Li Liu

6 indexed papers

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

Publications per year

6
26

Top categories

Crypto×4AI×2Neural Computing×1Vision×1Sound×1

Frequent co-authors

Zekun Fei2×
Ruiqi He2×
Zheli Liu2×
Weilin Lin1×
Ziqi Lin1×
Zhenxing Zhou1×

Research Timeline

2026
Not All Entities are Created Equal: A Dynamic Anonymization Framework for Privacy-Preserving RAG

The paper proposes TRIP-RAG, a dynamic anonymization framework that selectively anonymizes sensitive entities in knowledge bases used for RAG, significantly improving utility while maintaining strong privacy protection.

Misrouter: Exploiting Routing Mechanisms for Input-Only Attacks on Mixture-of-Experts LLMs

Misrouter introduces an input-only adversarial framework to exploit the routing mechanisms of Mixture-of-Experts (MoE) LLMs, enabling unsafe behavior induction against remotely hosted, black-box services.

Acoustic Interference: A New Paradigm Weaponizing Acoustic Latent Semantic for Universal Jailbreak against Large Audio Language Models

The paper introduces Acoustic Interference Attack (AIA), a novel jailbreak method that bypasses Large Audio Language Model (LALM) safety alignments by manipulating the underlying acoustic latent semantics rather than injecting malicious content.

Geodesics with Unified Tangent-constrained Priors and Curvature Regularization

The paper proposes a unified geodesic framework that combines tangent-constrained priors with curvature regularization to improve the robustness of image segmentation, especially for complex shapes.

Signed Spiking Neuron Enabled by an Orthogonal-Easy-Axis Magnetic Tunnel Junction

The paper proposes a compact magnetic tunnel junction (MTJ) device with orthogonal easy axes to implement signed leaky integrate-and-fire (LIF) neurons, enabling bipolar spike generation for enhanced neural network computation.

RedEdit: Agentic Red-Teaming of Image Safety Classifiers via MCTS-Guided Photo-Editing

The paper introduces RedEdit, an agentic red-teaming framework that demonstrates that malicious images can be easily edited to bypass safety classifiers while retaining their harmful semantics.

Highlighted terms show continued research focus across papers

Papers

cs.CRRecentJun 4, 2026

RedEdit: Agentic Red-Teaming of Image Safety Classifiers via MCTS-Guided Photo-Editing

Weilin Lin, Ziqi Lin, Zhenxing Zhou, Jianze Li +3 more

The paper introduces RedEdit, an agentic red-teaming framework that demonstrates that malicious images can be easily edited to bypass safety classifiers while retaining their harmful semantics.

View →
cs.NEcs.AIRecentJun 2, 2026

Signed Spiking Neuron Enabled by an Orthogonal-Easy-Axis Magnetic Tunnel Junction

Huannan Zheng, Jingli Liu, Kezhou Yang

The paper proposes a compact magnetic tunnel junction (MTJ) device with orthogonal easy axes to implement signed leaky integrate-and-fire (LIF) neurons, enabling bipolar spike generation for enhanced…

View →
cs.CVcs.AIRecentMay 28, 2026

Geodesics with Unified Tangent-constrained Priors and Curvature Regularization

Chong Di, Li Liu, Jinglin Zhang, Zhenjiang Li +2 more

The paper proposes a unified geodesic framework that combines tangent-constrained priors with curvature regularization to improve the robustness of image segmentation, especially for complex shapes.

View →
cs.CRcs.SDRecentMay 18, 2026

Acoustic Interference: A New Paradigm Weaponizing Acoustic Latent Semantic for Universal Jailbreak against Large Audio Language Models

Yanyun Wang, Yu Huang, Zi Liang, Xixin Wu +1 more

The paper introduces Acoustic Interference Attack (AIA), a novel jailbreak method that bypasses Large Audio Language Model (LALM) safety alignments by manipulating the underlying acoustic latent seman…

View →
cs.CRRecentMay 6, 2026

Misrouter: Exploiting Routing Mechanisms for Input-Only Attacks on Mixture-of-Experts LLMs

Zekun Fei, Zihao Wang, Weijie Liu, Ruiqi He +3 more

Misrouter introduces an input-only adversarial framework to exploit the routing mechanisms of Mixture-of-Experts (MoE) LLMs, enabling unsafe behavior induction against remotely hosted, black-box servi…

View →
cs.CRRecentMar 27, 2026

Not All Entities are Created Equal: A Dynamic Anonymization Framework for Privacy-Preserving RAG

Xinyuan Zhu, Zekun Fei, Enye Wang, Ruiqi He +4 more

The paper proposes TRIP-RAG, a dynamic anonymization framework that selectively anonymizes sensitive entities in knowledge bases used for RAG, significantly improving utility while maintaining strong…

View →