171 citations · 240 across the 9 of their papers we have counts for
9 papers
On Hallucination and Predictive Uncertainty in Conditional Language Generation
Yijun Xiao, William Yang Wang
Despite improvements in performances on different natural language generation tasks, deep neural models are prone to hallucinating facts that are incorrect or nonexistent. Differen…
Why Neural Machine Translation Prefers Empty Outputs
Xing Shi, Yijun Xiao, Kevin Knight
We investigate why neural machine translation (NMT) systems assign high probability to empty translations. We find two explanations. First, label smoothing makes correct-length tra…
Disentangled Representation Learning with Wasserstein Total Correlation
Yijun Xiao, William Yang Wang
Unsupervised learning of disentangled representations involves uncovering of different factors of variations that contribute to the data generation process. Total correlation penal…
Text Modeling with Syntax-Aware Variational Autoencoders
Yijun Xiao, William Yang Wang
Syntactic information contains structures and rules about how text sentences are arranged. Incorporating syntax into text modeling methods can potentially benefit both representati…
Quantifying Uncertainties in Natural Language Processing Tasks
Yijun Xiao, William Yang Wang
Reliable uncertainty quantification is a first step towards building explainable, transparent, and accountable artificial intelligent systems. Recent progress in Bayesian deep lear…
Implicit Regularization of Stochastic Gradient Descent in Natural Language Processing: Observations and Implications
Deren Lei, Zichen Sun, Yijun Xiao +1
Deep neural networks with remarkably strong generalization performances are usually over-parameterized. Despite explicit regularization strategies are used for practitioners to avo…
Dirichlet Variational Autoencoder for Text Modeling
Yijun Xiao, Tiancheng Zhao, William Yang Wang
We introduce an improved variational autoencoder (VAE) for text modeling with topic information explicitly modeled as a Dirichlet latent variable. By providing the proposed model t…
Tree-Structured Neural Machine for Linguistics-Aware Sentence Generation
Ganbin Zhou, Ping Luo, Rongyu Cao +4
Different from other sequential data, sentences in natural language are structured by linguistic grammars. Previous generative conversational models with chain-structured decoder i…
Efficient Character-level Document Classification by Combining Convolution and Recurrent Layers
Yijun Xiao, Kyunghyun Cho
Document classification tasks were primarily tackled at word level. Recent research that works with character-level inputs shows several benefits over word-level approaches such as…