Search Results for author:
Found 10 papers, 0 papers with code
Directional FDR Control for Sub-Gaussian Sparse GLMs
Using the debiased estimator, we establish multiple testing procedures.
On Hallucination and Predictive Uncertainty in Conditional Language Generation
Despite improvements in performances on different natural language generation tasks, deep neural models are prone to hallucinating facts that are incorrect or nonexistent.
Why Neural Machine Translation Prefers Empty Outputs
Using different EoS types in target sentences of different lengths exposes and eliminates this implicit smoothing.
Disentangled Representation Learning with Wasserstein Total Correlation
However, Kullback-Leibler (KL) divergence-based total correlation is metric-agnostic and sensitive to data samples.
Text Modeling with Syntax-Aware Variational Autoencoders
We propose syntax-aware variational autoencoders (SAVAEs) that dedicate a subspace in the latent dimensions dubbed syntactic latent to represent syntactic structures of sentences.
Quantifying Uncertainties in Natural Language Processing Tasks
Reliable uncertainty quantification is a first step towards building explainable, transparent, and accountable artificial intelligent systems.
Implicit Regularization of Stochastic Gradient Descent in Natural Language Processing: Observations and Implications
To bridge this gap, we study the role of SGD implicit regularization in deep learning systems.
Dirichlet Variational Autoencoder for Text Modeling
We introduce an improved variational autoencoder (VAE) for text modeling with topic information explicitly modeled as a Dirichlet latent variable.
Tree-Structured Neural Machine for Linguistics-Aware Sentence Generation
Then, with a proposed tree-structured search method, the model is able to generate the most probable responses in the form of dependency trees, which are finally flattened into sequences as the system output.
Efficient Character-level Document Classification by Combining Convolution and Recurrent Layers
Document classification tasks were primarily tackled at word level.
