Dessì, RobertoGualdoni, EleonoraFranzon, FrancescaBoleda, GemmaBaroni, Marco2023-01-242023-01-242022Dessi R, Gualdoni E, Franzon F, Boleda G, Baroni M. 2022. Communication breakdown: on the low mutual intelligibility between human and neural captioning. In: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022). 2022 dec 7-11; Abu Dhabi, United Arab Emirates. Stroudsburg: ACL, 2022. p. 7998–8007.http://hdl.handle.net/10230/55388Comunicació presentada a la 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022), celebrada del 7 a l'11 de desembre de 2022 a Abu Dhabi, Emirats Àrabs Units.We compare the 0-shot performance of a neural caption-based image retriever when given as input either human-produced captions or captions generated by a neural captioner. We conduct this comparison on the recently introduced IMAGECODE data-set (Krojer et al., 2022), which contains hard distractors nearly identical to the images to be retrieved. We find that the neural retriever has much higher performance when fed neural rather than human captions, despite the fact that the former, unlike the latter, were generated without awareness of the distractors that make the task hard. Even more remarkably, when the same neural captions are given to human subjects, their retrieval performance is almost at chance level. Our results thus add to the growing body of evidence that, even when the “language” of neural models resembles English, this superficial resemblance might be deeply misleading.application/pdfeng© ACL, Creative Commons Attribution 4.0 LicenseNeural modelsModels neuronalsinfo:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/openAccess