PUNC scores text-to-image uncertainty by comparing the prompt with an LVLM caption of the generated image, and this text-space similarity outperforms image-space baselines for OOD and ambiguous prompts.
Detecting Out- Of-Distribution Earth Observation Images with Diffusion Models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Towards Understanding and Quantifying Uncertainty for Text-to-Image Generation
PUNC scores text-to-image uncertainty by comparing the prompt with an LVLM caption of the generated image, and this text-space similarity outperforms image-space baselines for OOD and ambiguous prompts.