Semantic Behavioral Watermarking: Paraphrase-Robust and Forgery-Resistant Provenance for LLM Agents

Paper recorded by Signals 4 on 2026-10-06 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-06 on arXiv · recorded by Signals 4 on 2026-10-07

Category: cs.AI · 人工智能 · first seen 2026-10-07

Abstract

Behavioral watermarking embeds an owner identifier in an LLM agent's high-level action choices, giving provenance without touching output tokens. Prior agent watermarks break in two ways. First, all three prior schemes bind the watermark to the exact action symbol, so renaming a tool desynchronizes decoding even when the observation is untouched; in AgentMark's own robustness test, paraphrasing th

Read on arXiv →