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 language | English |
---|---|
Title of host publication | BlackboxNLP 2022 - BlackboxNLP Analyzing and Interpreting Neural Networks for NLP, Proceedings of the Workshop |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 325-334 |
Number of pages | 10 |
ISBN (Electronic) | 9781959429050 |
State | Published - 2022 |
Event | 5th 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
Name | BlackboxNLP 2022 - BlackboxNLP Analyzing and Interpreting Neural Networks for NLP, Proceedings of the Workshop |
---|
Conference
Conference | 5th 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/Territory | United Arab Emirates |
City | Abu Dhabi |
Period | 8/12/22 → … |
Bibliographical note
Publisher Copyright:© 2022 Association for Computational Linguistics.
Funding
We thank Carlo Meloni for his valuable feedback. This project received funding from the Europoean Research Council (ERC) under the Europoean Union’s Horizon 2020 research and innovation programme, grant agreement No. 802774 (iEXTRACT).
Funders | Funder number |
---|---|
Europoean Union’s Horizon 2020 research and innovation programme | 802774 |
European Commission |