Visually Guided Decoding: Gradient-Free Hard Prompt Inversion with Language Models.

Donghoon Kim, Minji Bae, Kyuhong Shim, Byonghyo Shim

Browse the full ICLR paper archive.