Who Is Behind the Harness? Fingerprinting LLMs through Agentic Behavior
cs.CR, cs.AI, cs.SE
Submitted: 2026-09-23
Updated: 2026-09-23
Code: https://github.com/SWE-agent/mini-swe-agent
Terminology
Sources
- TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories
- Sequential Behavioral Watermarking for LLM Agents
- ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
- One Token Is Enough: Fingerprinting and Verifying Large Language Models from Single-Token Output Distributions
- Did the Model Change? Efficiently Assessing Machine Learning API Shifts
- Agent Guide: A Simple Agent Behavioral Watermarking Framework
- Kimi K2: Open Agentic Intelligence
- DeepSeek-V3 Technical Report
- Watermarking LLM Agent Trajectories
- Agent trajectories as programs: fingerprinting and programming coding-agent behavior
- Qwen3 Technical Report
- Your "Pro" LLM Subscription May Actually Be "Free": Exposing Fingerprint Spoofing Risks in LLM Inference Services
- CoTSRF: Utilize Chain of Thought as Stealthy and Robust Fingerprint of Large Language Models
- SoK: Large Language Model Copyright Auditing via Fingerprinting
- Auditing Black-Box LLM APIs with a Rank-Based Uniformity Test
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