Sungroh Yoon

Building a Neural Machine Translation System Using Only Synthetic Parallel Data

2017 · arXiv (Cornell University)

Recent works have shown that synthetic parallel data automatically generated by translation models can be effective for various neural machine translation (NMT) issues. In this study, we build NMT systems using only synthetic parallel data. …

HexaGAN: Generative Adversarial Nets for Real World Classification

2019 · arXiv (Cornell University)

Most deep learning classification studies assume clean data. However, when dealing with the real world data, we encounter three problems such as 1) missing data, 2) class imbalance, and 3) missing label problems. These problems …

Interpretation of NLP models through input marginalization

2020 · arXiv (Cornell University)

To demystify the "black box" property of deep neural networks for natural language processing (NLP), several methods have been proposed to interpret their predictions by measuring the change in prediction probability after erasing each token …

Gradient Masking of Label Smoothing in Adversarial Robustness

2020 · IEEE Access

Deep neural networks (DNNs) have achieved impressive results in several image classification tasks. However, these architectures are unstable for adversarial examples (AEs) such as inputs crafted by a hardly perceptible perturbation with the intent of …

Rare Words Degenerate All Words.

2021 · arXiv (Cornell University)

Despite advances in neural network language model, the representation degeneration problem of embeddings is still challenging. Recent studies have found that the learned output embeddings are degenerated into a narrow-cone distribution which makes the similarity …