DALLE-2 is Seeing Double: Flaws in Word-to-Concept Mapping in Text2Image Models

Royi Rassin, Shauli Ravfogel, Yoav Goldberg

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

We study the way DALLE-2 maps symbols (words) in the prompt to their references (entities or properties of entities in the generated image). We show that in stark contrast to the way human process language, DALLE-2 does not follow the constraint that each word has a single role in the interpretation, and sometimes re-uses the same symbol for different purposes. We collect a set of stimuli that reflect the phenomenon: we show that DALLE-2 depicts both senses of nouns with multiple senses at once; and that a given word can modify the properties of two distinct entities in the image, or can be depicted as one object and also modify the properties of another object, creating a semantic leakage of properties between entities. Taken together, our study highlights the differences between DALLE-2 and human language processing and opens an avenue for future study on the inductive biases of text-to-image models.

Original languageEnglish
Title of host publicationBlackboxNLP 2022 - BlackboxNLP Analyzing and Interpreting Neural Networks for NLP, Proceedings of the Workshop
PublisherAssociation for Computational Linguistics (ACL)
Pages325-334
Number of pages10
ISBN (Electronic)9781959429050
StatePublished - 2022
Event5th Workshop on Analyzing and Interpreting Neural Networks for NLP, BlackboxNLP 2022 hosted by the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022 - Abu Dhabi, United Arab Emirates
Duration: 8 Dec 2022 → …

Publication series

NameBlackboxNLP 2022 - BlackboxNLP Analyzing and Interpreting Neural Networks for NLP, Proceedings of the Workshop

Conference

Conference5th Workshop on Analyzing and Interpreting Neural Networks for NLP, BlackboxNLP 2022 hosted by the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022
Country/TerritoryUnited Arab Emirates
CityAbu Dhabi
Period8/12/22 → …

Bibliographical note

Publisher Copyright:
© 2022 Association for Computational Linguistics.

Fingerprint

Dive into the research topics of 'DALLE-2 is Seeing Double: Flaws in Word-to-Concept Mapping in Text2Image Models'. Together they form a unique fingerprint.

Cite this