- Publications
- Influence
Language model pre-training, such as BERT, has significantly improved the performances of many natural language processing tasks. However, pre-trained language models are usually computationally… Continue Reading
We introduce multi-modal, attention-based neural machine translation (NMT) models which incorporate visual features into different parts of both the encoder and the decoder. We utilise global image… Continue Reading
In natural language generation (NLG), the task is to generate utterances from a more abstract input, such as structured data. An added challenge is to generate utterances that contain an accurate… Continue Reading
Neural Machine Translation (NMT) generates target words sequentially in the way of predicting the next word conditioned on the context words. At training time, it predicts with the ground truth words… Continue Reading
Paraphrasing exists at different granularity levels, such as lexical level, phrasal level and sentential level. This paper presents Decomposable Neural Paraphrase Generator (DNPG), a… Continue Reading
In this paper, we present an award-wining software environment for groundwater education and research. The software functions as a “digital laboratory” in which students can freely explore: visually… Continue Reading
IGW is a software environment for real-time 2D and 3D groundwater modeling. The software functions as a ''numerical laboratory'' in which the modeler can freely explore: visually creating an aquifer… Continue Reading
In this paper, we investigate the effectiveness of training a multimodal neural machine translation (MNMT) system with image features for a lowresource language pair, Hindi and English, using… Continue Reading
The pre-trained language models have achieved great successes in various natural language understanding (NLU) tasks due to its capacity to capture the deep contextualized information in text by… Continue Reading
Complex node interactions are common in knowledge graphs, and these interactions also contain rich knowledge information. However, traditional methods usually treat a triple as a training unit during… Continue Reading