HIVE: Harnessing Human Feedback for Instructional Visual Editing

arXiv:2303.09618v2 Announce Type: replace-cross
Abstract: Incorporating human feedback has been shown to be crucial to align text generated by large language models to human preferences. We hypothesize that state-of-the-art instructional image editing models, where outputs are generated based on an input image and an editing instruction, could similarly benefit from human feedback, as their outputs may not adhere to the correct instructions and preferences of users. In this paper, we present a novel framework to harness human feedback for instructional …

abstract art arxiv benefit cs.ai cs.cl cs.cv cs.hc cs.lg editing feedback generated hive human human feedback image language language models large language large language models state text type visual