Preprint Open access
Mutually Adversarial Self-Training with Evolving Data for Unified Multimodal Models
Unified multimodal models (UMMs) combine image generation and visual understanding in a shared backbone. Since generation and understanding are inverse tasks, recent studies self-train UMMs by letting the two branches cooperatively supervise each other. We introduce MATE (Mutually Adversarial self-Training with Evolvin …