SRAF: Stealthy and Robust Adversarial Fingerprint for Copyright Verification of Large Language Models
cs.CR
Submitted: 2025-05-08
Updated: 2026-09-18
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Functional Invariants to Watermark Large Transformers
- Invariant-based Robust Weights Watermark for Large Language Models
- Intrinsic Fingerprint of LLMs: Continue Training is NOT All You Need to Steal A Model!
- Improving Search Agent with One Line of Code
- CoTSRF: Utilize Chain of Thought as Stealthy and Robust Fingerprint of Large Language Models
- Invisible Traces: Using Hybrid Fingerprinting to identify underlying LLMs in GenAI Apps
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Detecting Stylistic Fingerprints of Large Language Models
- ImF: Implicit Fingerprint for Large Language Models
- DuFFin: A Dual-Level Fingerprinting Framework for LLMs IP Protection
- Gradient-Based Model Fingerprinting for LLM Similarity Detection and Family Classification
- LLM DNA: Tracing Model Evolution via Functional Representations
- AttnDiff: Attention-based Differential Fingerprinting for Large Language Models
- AdaMARP: An Adaptive Multi-Agent Interaction Framework for General Immersive Role-Playing
- Universal and Transferable Adversarial Attacks on Aligned Language Models
- Copyright Protection for Large Language Models: A Survey of Methods, Challenges, and Trends
- DNF: Dual-Layer Nested Fingerprinting for Large Language Model Intellectual Property Protection
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