AuxMark: Defending Against Unauthorized Agent Distillation via Auxiliary Behavioral Watermarking

arXiv:2609.34597 · cs.CR · Submitted 2026-09-28 · Read on arXiv

cs.CR

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/sierra-research/tau2-bench

Terminology

• Distillation — 66× in this paper · explained in 2 episodes
• Trace — 47× in this paper · explained in 2 episodes
• Core — 45× in this paper · explained in 2 episodes
• Robustness — 26× in this paper · explained in 5 episodes
• Accuracy — 15× in this paper · explained in 1 episode
• Verification — 15× in this paper · explained in 2 episodes
• Benchmark — 7× in this paper · explained in 2 episodes
• Generative AI — 4× in this paper · explained in 2 episodes
• Reasoning Traces — 4× in this paper · explained in 3 episodes
• Adam — 3× in this paper · explained in 2 episodes
• Aggregation — 2× in this paper · explained in 1 episode
• Consistency — 2× in this paper · explained in 3 episodes
• Cosine Similarity — 2× in this paper · explained in 3 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Fingerprinting — 2× in this paper · explained in 2 episodes
• Hallucination — 2× in this paper · explained in 3 episodes
• Homogenization — 2× in this paper · explained in 2 episodes
• Action Graph Similarity — 1× in this paper
• Adversarial Attacks — 1× in this paper · explained in 1 episode
• Agent Harness — 1× in this paper · explained in 2 episodes
• Agent Interactions — 1× in this paper · explained in 1 episode
• Agentic Evaluation — 1× in this paper · explained in 1 episode
• Agentic System — 1× in this paper · explained in 2 episodes
• Knowledge Distillation — 1× in this paper · explained in 11 episodes
• LLM judge — 1× in this paper · explained in 2 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• LoRA Adapters — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• gpt-oss-120b & gpt-oss-20b Model Card
• Sequential Behavioral Watermarking for LLM Agents
• The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence
• Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
• UltraFeedback: Boosting Language Models with Scaled AI Feedback
• ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
• The False Promise of Imitating Proprietary LLMs
• Distilling the Knowledge in a Neural Network
• Agent Guide: A Simple Agent Behavioral Watermarking Framework
• SAMark: A Self-Anchored Text Watermarking with Paragraph-Level Paraphrase Robustness
• Ministral 3
• An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Traffic
• AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent
• Student-Centered Distillation Narrows the Agentic Gap Between Small and Large LLMs
• Orca: Progressive Learning from Complex Explanation Traces of GPT-4
• Stealing Reasoning Traces from Proprietary LLM APIs
• Reference-Based Distillation Detection in LLMs
• Artificial Intelligence Index Report 2026
• OpenAI GPT-5 System Card
• Kimi K2: Open Agentic Intelligence

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills
• Token Composition: A Graph Based on EVM Logs