SOURCE-LINKED INTELLIGENCE

EvoHarmBench: Breaking Content Moderation with Iterative Human-Like Evasion

arXiv · AI, language, vision and robotics · article · Aug 28, 2026 · UTC

Existing evaluations of harmful content detection rely predominantly on static benchmarks, which struggle to reflect the interactive adversarial ecosystem of real-world content platforms where users continuously revise their expressions in response to moderation feedback. This mismatch creates a significant performance gap between offline benchmark scores and online deployment effectiveness. To the best of our knowledge, we present EvoHarmBench, the first dynamic adversarial evaluation framework for content moderation systems. The framework employs an iterative optimization loop that evolves e

Read original source ↗ Open in workspace

recordType

paper

region

Global