Discovering the Hidden Vocabulary of DALLE-2. (arXiv:2206.00169v1 [cs.LG])

We discover that DALLE-2 seems to have a hidden vocabulary that can be used
to generate images with absurd prompts. For example, it seems that
\texttt{Apoploe vesrreaitais} means birds and \texttt{Contarra ccetnxniams
luryca tanniounons} (sometimes) means bugs or pests. We find that these prompts
are often consistent in isolation but also sometimes in combinations. We
present our black-box method to discover words that seem random but have some
correspondence to visual concepts. This creates important security and
interpretability challenges.

arxiv dalle dalle-2