Self-guided curriculum learning for neural machine translation
In the field of machine learning, the well-trained model is assumed to be able to recover the
training labels, ie the synthetic labels predicted by the model should be as close to the …
training labels, ie the synthetic labels predicted by the model should be as close to the …
Can Linguistic Knowledge Improve Multimodal Alignment in Vision-Language Pretraining?
The multimedia community has shown a significant interest in perceiving and representing
the physical world with multimodal pretrained neural network models, and among them, the …
the physical world with multimodal pretrained neural network models, and among them, the …
A crosslingual investigation of conceptualization in 1335 languages
Languages differ in how they divide up the world into concepts and words; eg, in contrast to
English, Swahili has a single concept forbelly'andwomb'. We investigate these differences in …
English, Swahili has a single concept forbelly'andwomb'. We investigate these differences in …
Bridging the gap between clean data training and real-world inference for spoken language understanding
Spoken language understanding (SLU) system usually consists of various pipeline
components, where each component heavily relies on the results of its upstream ones. For …
components, where each component heavily relies on the results of its upstream ones. For …
Speech Sense Disambiguation: Tackling Homophone Ambiguity in End-to-End Speech Translation
End-to-end speech translation (ST) presents notable disambiguation challenges as it
necessitates simultaneous cross-modal and cross-lingual transformations. While word …
necessitates simultaneous cross-modal and cross-lingual transformations. While word …
Evalign: Visual evaluation of translation alignment models
This paper presents EvAlign, a visual analytics framework for quantitative and qualitative
evaluation of automatic translation alignment models. EvAlign offers various visualization …
evaluation of automatic translation alignment models. EvAlign offers various visualization …
Light weight IBP deep residual network for image super resolution
H Lin, J Yang - IEEE Access, 2021 - ieeexplore.ieee.org
Single-image super resolution (SR) is used to reconstruct a high-resolution image with more
high-frequency details based on a low-resolution image as input. In recent years, image SR …
high-frequency details based on a low-resolution image as input. In recent years, image SR …
LaDA: Latent Dialogue Action For Zero-shot Cross-lingual Neural Network Language Modeling
Z Ma, J Ye, S Cheng - arXiv preprint arXiv:2308.02903, 2023 - arxiv.org
Cross-lingual adaptation has proven effective in spoken language understanding (SLU)
systems with limited resources. Existing methods are frequently unsatisfactory for intent …
systems with limited resources. Existing methods are frequently unsatisfactory for intent …
Noisy Parallel Data Alignment
R Xie, A Anastasopoulos - arXiv preprint arXiv:2301.09685, 2023 - arxiv.org
An ongoing challenge in current natural language processing is how its major
advancements tend to disproportionately favor resource-rich languages, leaving a …
advancements tend to disproportionately favor resource-rich languages, leaving a …
Leveraging neural machine translation for word alignment
V Zouhar, D Pylypenko - arXiv preprint arXiv:2103.17250, 2021 - arxiv.org
The most common tools for word-alignment rely on a large amount of parallel sentences,
which are then usually processed according to one of the IBM model algorithms. The …
which are then usually processed according to one of the IBM model algorithms. The …