Neural Machine Translation with Universal Visual Representation

122 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Multimodal Chain-of-Thought Reasoning in Language Models (2023, neutral): “After obtaining the patch-level vision features, we apply a learnable projection matrix W h to convert the shape of VisionExtractor ( X vision ) into that of H language ; thus we have H vision ∈ R m × d where m is the number of patches.”
Multilingual Multimodal Pre-training for Zero-Shot Cross-Lingual Transfer of Vision-Language Models (2021, background): “While spoken languages could be different, all humans share similar vision systems, and many visual concepts can be understood universally (Sigurdsson et al., 2020; Zhang et al., 2020).”
Good for Misconceived Reasons: An Empirical Revisiting on the Need for Visual Context in Multimodal Machine Translation (2021, methodology): “We follow (Zhang et al., 2020) to early-stop the training if validation loss does not improve for ten epochs.”
On Vision Features in Multimodal Machine Translation (2022, methodology): “We also adopted the early-stop training strategy (Zhang et al., 2020) to avoid the overﬁtting issue.”
Universal Multimodal Representation for Language Understanding (2023, methodology): “Part of this study has been published as “Neural Machine Translation with Universal Visual Representation” [37] in the Eighth International Conference on Learning Representations (ICLR 2020).”