EVADE-Bench: Multimodal Benchmark for Evaluating and Enhancing Evasive Content Detection 文章

arXiv:2505.17654v4 Announce Type: replace Abstract: E-commerce platforms increasingly rely on Large Language Models (LLMs) and Vision Language Models (VLMs) to detect illicit or misleading product content. However, these models remain vulnerable to evasive content, which refers to inputs that have been deliberately modified through techniques such as word splitting, euphemistic language, or image cropping to conceal policy violations while still conveying prohibited claims. Crucially, detecting such content requires a model to simultaneously master two capabilities: accurately comprehending complex rules, and correctly inferring the true intent behind deliberately obfuscated multimodal inputs. While prior work has separately explored LLM reasoning over complex rules and LLM-based detection of evasive content, no existing benchmark combines both within a unified evaluation framework.