Word Representation Learning in Multimodal Pre-Trained Transformers: An Intrinsic Evaluation
AbstractThis study carries out a systematic intrinsic evaluation of the semantic representations learned by state-of-the-art pre-trained multimodal Transformers. These representations are claimed to be task-agnostic and shown to help on many downstream language-and-vision tasks. Howe...
| Asıl Yazarlar: | , , |
|---|---|
| Materyal Türü: | Makale |
| Dil: | English |
| Baskı/Yayın Bilgisi: |
The MIT Press
2021-01-01
|
| Seri Bilgileri: | Transactions of the Association for Computational Linguistics |
| Online Erişim: | https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00443/108935/Word-Representation-Learning-in-Multimodal-Pre |