AI RESEARCH
TwoHamsters: Benchmarking Multi-Concept Compositional Unsafety in Text-to-Image Models
arXiv CS.CV
•
ArXi:2604.15967v1 Announce Type: cross Despite the remarkable synthesis capabilities of text-to-image (T2I) models, safeguarding them against content violations remains a persistent challenge. Existing safety alignments primarily focus on explicit malicious concepts, often overlooking the subtle yet critical risks of compositional semantics. To address this oversight, we identify and formalize a novel vulnerability: Multi-Concept Compositional Unsafety (MCCU), where unsafe semantics stem from the implicit associations of individually benign concepts. Based on this formulation, we.