We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Machine Learning

Title: Discovering the Hidden Vocabulary of DALLE-2

Abstract: 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.
Comments: 6 pages, 4 figures
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2206.00169 [cs.LG]
  (or arXiv:2206.00169v1 [cs.LG] for this version)

Submission history

From: Giannis Daras [view email]
[v1] Wed, 1 Jun 2022 01:14:48 GMT (21335kb,D)

Link back to: arXiv, form interface, contact.