Pith. sign in

REVIEW 3 cited by

Discovering the Hidden Vocabulary of DALLE-2

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2206.00169 v1 pith:2XMOELAS submitted 2022-06-01 cs.LG cs.CLcs.CRcs.CV

Giannis Daras , Alexandros G. Dimakis This is my paper
classification cs.LGcs.CLcs.CRcs.CV
keywords dalle-2discoverhiddenmeanspromptsseemssometimestexttt
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original 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.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety

    cs.CY 2026-06 accept novelty 6.5 of 10

    Legal and ethical bans on CSAM access and generation break standard AI safety techniques, creating 15 open problems that demand new methods for dataset cleaning, concept fusion prevention, fine-tuning resilience, dete...

  2. Mind the Gap: Zero-Query Jailbreaks via Filter-Generator Discrepancy in Text-to-Image Systems

    cs.CL 2026-08 conditional novelty 6.0 of 10

    Zero-query jailbreak attacks on text-to-image systems that exploit filter-generator discrepancy reach 29-33% average success and beat baselines on six pipelines and GPT-image-2.

  3. Rethinking Layered Graphic Design Generation with a Top-Down Approach

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Accordion decomposes AI-generated raster designs into editable background, object, and vectorized text layers using a VLM-driven top-down planning pipeline.

Pith tools