Papers arxiv:2605.30501 Linear Ensembles Wash Away Watermarks: On the Fragility of Distributional Perturbations in LLMs Published on May 28 Submitted by Zhihao Wu on Jun 2 King's College London Authors: Zhihao Wu , , , Yudong Chen , Runcong Zhao Abstract Watermarking AI-generated text for detection… …
… This design is grounded in our theoretical framework that characterizes a jointly optimal embedding-detection pair, achieving the fundamental trade-offs among detection accuracy , robustness , and distortion . …
… Evaluation on four QA datasets and four attack types demonstrates the effectiveness of spectral entropy for backdoor detection. …
… Its standalone watermarks further achieve 93.6–100.0% detection with at most 1.9% false alarms . 🚀 Agent traces are not just logs — they are security interfaces. 🌐 Project: https://xushuwenn.github.io/RedAct Website/ 💻 Code: https://github.com/XuShuwenn/RedAct 📄 Paper: https://arxiv.org/abs/2606.10… …