5 papers
Agent Skills Matter: Inferring Proprietary Skills from Execution Trajectories
Jianing Geng, Ruiqi He, Zekun Fei +6
Agent skills package reusable procedures that improve downstream performance. Their lightweight, portable form enables marketplace monetization and private deployment behind cloud-…
Misrouter: Exploiting Routing Mechanisms for Input-Only Attacks on Mixture-of-Experts LLMs
Zekun Fei, Zihao Wang, Weijie Liu +4
Mixture-of-Experts (MoE) architectures have emerged as a leading paradigm for scaling large language models through sparse, routing-based computation. However, this design introduc…
Hiding in Plain Sight: A Steganographic Approach to Stealthy LLM Jailbreaks
Jianing Geng, Biao Yi, Zekun Fei +4
Jailbreak attacks pose a serious threat to Large Language Models (LLMs) by bypassing their safety mechanisms. A truly advanced jailbreak is defined not only by its effectiveness bu…
BadReasoner: Planting Tunable Overthinking Backdoors into Large Reasoning Models for Fun or Profit
Biao Yi, Zekun Fei, Jianing Geng +4
Large reasoning models (LRMs) have emerged as a significant advancement in artificial intelligence, representing a specialized class of large language models (LLMs) designed to tac…
Your Semantic-Independent Watermark is Fragile: A Semantic Perturbation Attack against EaaS Watermark
Zekun Fei, Biao Yi, Jianing Geng +3
Embedding-as-a-Service (EaaS) has emerged as a successful business pattern but faces significant challenges related to various forms of copyright infringement, particularly, the AP…