AI RESEARCH
When Identities Collapse: A Stress-Test Benchmark for Multi-Subject Personalization
arXiv CS.AI
•
ArXi:2603.26078v1 Announce Type: cross Subject-driven text-to-image diffusion models have achieved remarkable success in preserving single identities, yet their ability to compose multiple interacting subjects remains largely unexplored and highly challenging. Existing evaluation protocols typically rely on global CLIP metrics, which are insensitive to local identity collapse and fail to capture the severity of multi-subject entanglement.