Abstract
Large language model agents can acquire complex capabilities through multi-step interaction and tool use, but their trajectories can also be illegally collected to dis- till student agents. However, existing watermarking methods either do not fit the structured and interactive nature of agent environments or lack reliable effective- ness across tasks and model architectures. We introduce AuxMark, a behavioral watermarking framework for tracing unauthorized agent distillation. AuxMark dynamically inserts safe, non-essential auxiliary action into teacher trajectories, and stores the associated contexts as private evidence cards. To audit a suspicious student model, AuxMark constructs paired real and fake probes from these cards and applies a card-level sign test. This black-box protocol supports both model- level detection and trace-level attribution. Across three agent benchmarks, two teacher agents, and four student architectures, AuxMark detects all 24 distilled models with zero false positives on 48 clean models. It also preserves task utility and remains effective against data flooding, paraphrasing, truncation, and adaptive cleaning attacks. Our code will be released at this URL.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Feng, Y., Feng, H., Shang, S., Zhang, X., Lou, J., Zhao, H., & Zhou, M. (2026). AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking. https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking
MLA 9
Feng, Yiqing, et al. "AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking." https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking.
Chicago (author–date)
Feng, Yiqing, Haozhe Feng, Shunan Shang, Xiaoyu Zhang, Jian Lou, Haodong Zhao, and Mingxun Zhou. 2026. "AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking." https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking.
Harvard
Feng, Y., Feng, H., Shang, S., Zhang, X., Lou, J., Zhao, H. and Zhou, M. (2026) 'AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking', Available at: https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking.
Vancouver
Feng Y, Feng H, Shang S, Zhang X, Lou J, Zhao H, et al. AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking. https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking
IEEE
Y. Feng, H. Feng, S. Shang, X. Zhang, J. Lou, H. Zhao, and M. Zhou, "AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking," https://omanscience.com/en/articles/auxmark-defending-against-unauthorized-agent-distillation-via-auxiliary-behavioral-watermarking.