I really hope we see improvements to the resolutions large multimodal models can handle. Right now this patchwork approach leads to lots of unwieldly workaround

I really hope we see improvements to the resolutions large multimodal models can handle. Right now this patchwork approach leads to lots of unwieldly workarounds in applications.