Semantic Behavioral Watermarking: Paraphrase-Robust and Forgery-Resistant Provenance for LLM Agents
cs.CR, cs.AI
Submitted: 2026-10-06
Updated: 2026-10-06
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- PASA: A Principled Embedding-Space Watermarking Approach for LLM-Generated Text under Semantic-Invariant Attacks
- Sequential Behavioral Watermarking for LLM Agents
- SemStamp: A Semantic Watermark with Paraphrastic Robustness for Text Generation
- k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated Text
- Agent Guide: A Simple Agent Behavioral Watermarking Framework
- AgentMark: Utility-Preserving Behavioral Watermarking for Agents
- Watermark Stealing in Large Language Models
- A Watermark for Large Language Models
- On the Reliability of Watermarks for Large Language Models
- Robust Distortion-free Watermarks for Language Models
- A Semantic Invariant Robust Watermark for Large Language Models
- Tool-Planner: Task Planning with Clusters across Multiple Tools
- APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasets
- Watermarking LLM Agent Trajectories
- When Only the Final Text Survives: Implicit Execution Tracing for Multi-Agent Auditing
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
- On Protecting Agentic Systems' Intellectual Property via Watermarking
- MemMark: State-Evolution Attribution Watermarking for Agent Long-Term Memory Systems
- xLAM: A Family of Large Action Models to Empower AI Agent Systems
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