EUVIMEDEUVIMED entdecken
Crossref

Transformer Module Networks for Systematic Generalization in Visual Question Answering

Moyuru Yamada, Vanessa D'Amario, Kentaro Takemoto, Xavier Boix, Tomotake Sasaki

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-12 · 46 (12) · 10096-10105

DOI: 10.1109/tpami.2024.3438887 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-12. Verfasst von Moyuru Yamada und 4 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 33 Literaturangaben. Gefördert von: Fujitsu Limited, NSF STC, National Institute of Advanced Industrial Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformers achieve great performance on Visual Question Answering (VQA). However, their systematic generalization capabilities, i.e., handling novel combinations of known concepts, is unclear. We reveal that Neural Module Networks (NMNs), i.e., question-specific compositions of modules that tackle a sub-task, achieve better or similar systematic generalization performance than the conventional Transformers, even though NMNs' modules are CNN-based. In order to address this shortcoming of Transformers with respect to NMNs, in this paper we investigate whether and how modularity can bring benefits to Transformers. Namely, we introduce Transformer Module Network (TMN), a novel NMN based on compositions of Transformer modules. TMNs achieve state-of-the-art systematic generalization performance in three VQA datasets, improving more than 30% over standard Transformers for novel compositions of sub-tasks. We show that not only the module composition but also the module specialization for each sub-task are the key of such performance gain.

Förderung

  • Fujitsu Limited
  • NSF STC
  • National Institute of Advanced Industrial Science and Technology

Zitieren

APA

Yamada, M., D'Amario, V., Takemoto, K., Boix, X., & Sasaki, T. (2024). Transformer Module Networks for Systematic Generalization in Visual Question Answering. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 10096-10105. https://doi.org/10.1109/tpami.2024.3438887

Vancouver

Yamada M, D'Amario V, Takemoto K, Boix X, Sasaki T. Transformer Module Networks for Systematic Generalization in Visual Question Answering. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(12):10096-10105. doi:10.1109/tpami.2024.3438887

BibTeX
@article{Yamada2024,
  author = {Yamada, Moyuru and D'Amario, Vanessa and Takemoto, Kentaro and Boix, Xavier and Sasaki, Tomotake},
  title = {Transformer Module Networks for Systematic Generalization in Visual Question Answering},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2024},
  volume = {46},
  number = {12},
  pages = {10096--10105},
  doi = {10.1109/tpami.2024.3438887}
}
Kostenlos starten →