arXiv Computation and Language By Ruijie Jian, Benlei Cui, Ting Ma, Haidong Ding, Kangwei Liu, Ziwen Xu, Longtao Huang, Hui Xue, Ziqiang Zhu, Junjie Li, Haiwen Hong

EvoHarmBench: Breaking Content Moderation with Iterative Human-Like Evasion

Read the original on arXiv Computation and Language →

EvoHarmBench is a dynamic adversarial evaluation framework that simulates how users iteratively modify harmful content to evade moderation. It uses an optimization loop that evolves evasion strategies at the semantic-cluster level while maintaining human readability, and tests 229 semantic sub-clusters across five violation categories derived from 5,002 real-world adversarial samples. The study shows that even state‑of‑the‑art LLM‑based moderators can be bypassed with an 80.3% success rate after twelve iterations, highlighting significant vulnerabilities in current systems.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI
Aug 3

TextCloak: Thwarting Unauthorized LLM Exploitation via RL-Driven Unlearnable Text

arXiv:2607. 28862v1 Announce Type: cross Abstract: The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage.

By Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu