MENTOR: Efficient Multimodal-Conditioned Tuning for Autoregressive Vision Generation Models 事件

MENTOR: Efficient Multimodal-Conditioned Tuning for Autoregressive Vision Generation Models arXiv:2507.09574v3 Announce Type: replace Abstract: Recent text-to-image models produce high-quality results but still struggle with precise visual control, balancing multimodal inputs, and requiring extensive training for complex multimodal image generation. To address these limitations, we propose MENTOR, a novel autoregressive (AR) framework for efficient Multimodal-conditioned Tuning for Autoregressi