Abstract: Inspired by the performance and scalability of autoregressive large language models (LLMs), transformer-based models have seen recent success in the visual domain. This study investigates a ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results