Publications (21)
Boosting Adversarial Transferability via High-Frequency Augmentation and Hierarchical-Gradient Fusion
Yayin Zheng, Chen Wan, Zihong Guo +2
Adversarial attacks have become a significant challenge in the security of machine learning models, particularly in the context of black-box defense strategies. Existing methods fo…
On the residue method for period integrals
Aaron Pollack, Chen Wan, MichaÅ Zydor
By applying the residue method for period integrals and Langlands-Shahidi's theory for residues of Eisenstein series, we study the period integrals for six spherical varieties. For…
Multiplicity One Theorem for the Ginzburg-Rallis Model: the tempered case
Chen Wan
Following the method developed by Waldspurger and Beuzart-Plessis in their proof of the local Gan-Gross-Prasad conjecture, we are able to prove the multiplicity one theorem for the…
BZSV Duality for Some Strongly Tempered Spherical Varieties
Zhengyu Mao, Chen Wan, Lei Zhang
We propose two families of relative trace formula comparisons in the study of relative Langlands duality conjectured by Ben-Zvi--Sakellaridis--Venkatesh. This allows us to incorpor…
A Two-Stage Globally-Diverse Adversarial Attack for Vision-Language Pre-training Models
Wutao Chen, Huaqin Zou, Chen Wan +1
Vision-language pre-training (VLP) models are vulnerable to adversarial examples, particularly in black-box scenarios. Existing multimodal attacks often suffer from limited perturb…
A Local Relative Trace Formula for the Ginzburg-Rallis Model: the Geometric Side
Chen Wan
Following the method developed by Waldspurger and Beuzart-Plessis in their proofs of the local Gan-Gross-Prasad conjecture, we are able to prove the geometric side of a local relat…