Author

Beining Xu

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Sep 2026

Backdoor in the Loop: Compromising Agentic Search via Malicious Retrievers

This work proposes a controlled inject-and-remove cycle: deliberately inject a weaker backdoor and then unlearn it, which weakens detector-visible signatures and fools the backdoor detectors with an illusion of purification while preserving the malicious retrieval behavior.

Bei-Ning Xu, Pei-Chun Hua, Yun-Ming Xiao · 0 citations

Preprint Aug 2026

Beyond Visual Evidence: Revealing and Mitigating Relational Privacy Leakage in Document MLLMs

The Dynamic Relational Unlearning Framework (DRUF) is proposed which comprises a Relational Decoupling Unlearning (RDU) module and a dynamic set update mechanism that suppresses the leakage of high-risk field pairs while preserving KIE performance.

Bei-Ning Xu, Hairui Wang, Jiaxin Wang et al. · 0 citations