AI RESEARCH
Dynamic Optimization and Safety Indicator Injection for Jailbreaking Text-to-Image Models with Multimodal Safety Filters
arXiv CS.LG
•
ArXi:2505.18979v2 Announce Type: replace Text-to-image (T2I) models can generate not-safe-for-work (NSFW) content, motivating multi-stage safety pipelines with both text and image filters. Newer LLM-based filters detect latent intent beyond keywords, making token-level perturbation attacks unreliable. Our evaluation further shows that existing jailbreak methods exhibit a sharp trade-off between filter evasion and semantic fidelity, while also requiring excessive queries to succeed. We