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

Feng Yang

4 indexed papers

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

Publications per year

4
26

Top categories

Vision×2Crypto×2Signal Processing×1Software Eng.×1

Frequent co-authors

Yang Liu2×
Hongpu Zhang1×
Shu Sun1×
Ruifeng Gao1×
Tongjia Zhang1×
Lianyu Hu1×

Research Timeline

2026
SHIFT: Stochastic Hidden-Trajectory Deflection for Removing Diffusion-based Watermark

The paper introduces SHIFT, a training-free attack that exploits the vulnerability of diffusion-based watermarking by stochastically deflecting the generative trajectory, achieving high removal rates across diverse watermarking methods.

Probing Privacy Leaks in LLM-based Code Generation via Test Generation

The paper proposes a novel test-driven pipeline that simulates realistic code generation scenarios to detect privacy leaks in LLMs, achieving a 2.56x increase in detected leakage compared to existing methods.

CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts

This paper proposes CoLT, a framework that enables multi-modal models to reason through a chain of latent thought representations instead of text tokens, improving performance and reducing inference time.

A Foundation Model for Cross-Band CSI Reconstruction

This paper proposes a model for cross-band CSI reconstruction in multi-band low-altitude wireless systems using radio-frequency metadata and pilot-guided cross-attention.

Highlighted terms show continued research focus across papers

Papers

eess.SPEmpiricalRecentJul 24, 2026

A Foundation Model for Cross-Band CSI Reconstruction

Hongpu Zhang, Shu Sun, Ruifeng Gao, Tongjia Zhang +1 more

This paper proposes a model for cross-band CSI reconstruction in multi-band low-altitude wireless systems using radio-frequency metadata and pilot-guided cross-attention.

View →
cs.CVEmpiricalRecent
Jun 30, 2026

CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts

Lianyu Hu, Shengqian Qin, Zeqin Liao, Qing Guo +3 more

This paper proposes CoLT, a framework that enables multi-modal models to reason through a chain of latent thought representations instead of text tokens, improving performance and reducing inference t…

View →
cs.SEcs.CRMay 14, 2026

Probing Privacy Leaks in LLM-based Code Generation via Test Generation

Yifei Ge, Zhenpeng Chen, Weisong Sun, Yuchen Chen +6 more

The paper proposes a novel test-driven pipeline that simulates realistic code generation scenarios to detect privacy leaks in LLMs, achieving a 2.56x increase in detected leakage compared to existing…

View →
cs.CVcs.CRMar 31, 2026

SHIFT: Stochastic Hidden-Trajectory Deflection for Removing Diffusion-based Watermark

Rui Bao, Zheng Gao, Xiaoyu Li, Xiaoyan Feng +2 more

The paper introduces SHIFT, a training-free attack that exploits the vulnerability of diffusion-based watermarking by stochastically deflecting the generative trajectory, achieving high removal rates…

View →