Skip to content

Author

Hao-Yu Wang

We have 5 of 14 papers

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.

Aug 2026

Not All Tokens Are Equal: Region-Aware Consistency Repair of Backdoors in MLLMs

RACER is presented, a model-level repair framework motivated by a key observation: backdoors induce abnormal layer-to-layer evolution in internal representations, which is modality-dependent, concentrating primarily in the token region encoding the trigger features that the backdoor model actually relies on.

Jia-Li Wei, Ming Fan, Ming-Kun Zhang et al. · 0 citations
#artificial intelligence Preprint Sep 2026

Representation Transitions Reveal Emerging Safety Risks in Multi-Turn LLM Agents

Multi-turn attacks on agentic systems can compose individually permissible actions into harmful outcomes, challenging defenses that assess actions or states in isolation. We show that such attacks leave a detectable signature in the agent's internal representations: harmful behavior emerges as an accumulated representa...

Hao-Yu Wang, Wei Zhao, Ye-Di Zhang et al. · 1 citation
#artificial intelligence Preprint Sep 2026

PolicyMem: Geometric Policy Memory for LLM Governance

PolicyMem is introduced, a geometric policy memory that externalizes natural-language policies as reusable geometric memory objects represented by low-rank subspaces in a shared representation space that achieves state-of-the-art unsafe behavior detection while enabling effective policy attribution, rewriting, and post...

Yuan-Chen Bei, Zheng-Zhang Chen, Yan-Jun Zhao et al. · 0 citations
Preprint Aug 2026

$S^3$: Improving Agent Safety through Multi-Stage Defense

An automated transformation pipeline that converts existing safety designs into reusable safety skills and establish a community-driven safety skill library is developed and results demonstrate the potential of stage-specific safety skills as a scalable and composable foundation for building resilient and trustworthy a...

Zi-Bo Xiao, Hao-Yu Wang, Jun Sun · 0 citations
Preprint Aug 2026

MemGuard: Persisting Verifier Signals for LLM-Agent Memory Governance

Averaged over five seeds, MemGuard achieves the best success metric and lowest average steps in all 16 backbone-benchmark settings, improving over ReasoningBank, the strongest prior baseline among the memory methods the authors evaluate.

Hao-Yu Wang, Guang-Yuan Dong, He Liang et al. · 4 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.