Anna Sun
YOU?
Author Swipe
View article: Genome-wide meta-analyses of cross substance use disorders in diverse populations
Genome-wide meta-analyses of cross substance use disorders in diverse populations Open
Substance use disorders (SUDs, including alcohol, cannabis, opioids, and tobacco) represent significant public health challenges. The estimated heritability of SUDs is ~50% and many individuals experience multiple SUDs concurrently. Studie…
View article: Real-world observations on neuroinflammation-related drug responses in Alzheimer's disease
Real-world observations on neuroinflammation-related drug responses in Alzheimer's disease Open
Background Alcohol use disorder (AUD), epilepsy, hemorrhagic stroke (HS), and traumatic brain injury (TBI) are all linked to neuroinflammation and associated with an increased risk of Alzheimer's disease (AD). Drug responses in cognitive h…
View article: Real‐world observations of GLP‐1 receptor agonists and SGLT‐2 inhibitors as potential treatments for Alzheimer's disease
Real‐world observations of GLP‐1 receptor agonists and SGLT‐2 inhibitors as potential treatments for Alzheimer's disease Open
INTRODUCTION Glucagon‐like peptide‐1 (GLP‐1) receptor agonists, sodium‐glucose cotransporter‐2 (SGLT‐2) inhibitors, and dipeptidyl peptidase‐4 (DPP‐4) inhibitors have potential beneficial effects in Alzheimer's disease (AD). METHODS We con…
View article: Human herpesvirus‐associated transposable element activation in human aging brains with Alzheimer's disease
Human herpesvirus‐associated transposable element activation in human aging brains with Alzheimer's disease Open
INTRODUCTION Human herpesvirus (HHV) has been linked to Alzheimer's disease (AD), but the underlying mechanisms remain unknown. METHODS We leveraged functional genomics data from Religious Orders Study or the Rush Memory and Aging Project …
View article: A Precision Mixture Risk Model to Identify Adverse Drug Events in Subpopulations Using a Case‐Crossover Design
A Precision Mixture Risk Model to Identify Adverse Drug Events in Subpopulations Using a Case‐Crossover Design Open
Despite the success of pharmacovigilance studies in detecting signals of adverse drug events (ADEs) from real‐world data, the risks of ADEs in subpopulations warrant increased scrutiny to prevent them in vulnerable individuals. Recently, t…
View article: MMM: Multi-Layer Multi-Residual Multi-Stream Discrete Speech Representation from Self-supervised Learning Model
MMM: Multi-Layer Multi-Residual Multi-Stream Discrete Speech Representation from Self-supervised Learning Model Open
Speech discrete representation has proven effective in various downstream applications due to its superior compression rate of the waveform, fast convergence during training, and compatibility with other modalities. Discrete units extracte…
View article: Associations Between Neuroinflammation-Related Conditions and Alzheimer’s Disease: A Study of US Insurance Claims Data
Associations Between Neuroinflammation-Related Conditions and Alzheimer’s Disease: A Study of US Insurance Claims Data Open
Background: Early detection of Alzheimer’s disease (AD) is a key component for the success of the recently approved lecanemab and aducanumab. Patients with neuroinflammation-related conditions are associated with a higher risk for developi…
View article: Seamless: Multilingual Expressive and Streaming Speech Translation
Seamless: Multilingual Expressive and Streaming Speech Translation Open
Large-scale automatic speech translation systems today lack key features that help machine-mediated communication feel seamless when compared to human-to-human dialogue. In this work, we introduce a family of models that enable end-to-end …
View article: Efficient Monotonic Multihead Attention
Efficient Monotonic Multihead Attention Open
We introduce the Efficient Monotonic Multihead Attention (EMMA), a state-of-the-art simultaneous translation model with numerically-stable and unbiased monotonic alignment estimation. In addition, we present improved training and inference…
View article: Multi-resolution HuBERT: Multi-resolution Speech Self-Supervised Learning with Masked Unit Prediction
Multi-resolution HuBERT: Multi-resolution Speech Self-Supervised Learning with Masked Unit Prediction Open
Existing Self-Supervised Learning (SSL) models for speech typically process speech signals at a fixed resolution of 20 milliseconds. This approach overlooks the varying informational content present at different resolutions in speech signa…
View article: SeamlessM4T: Massively Multilingual & Multimodal Machine Translation
SeamlessM4T: Massively Multilingual & Multimodal Machine Translation Open
What does it take to create the Babel Fish, a tool that can help individuals translate speech between any two languages? While recent breakthroughs in text-based models have pushed machine translation coverage beyond 200 languages, unified…
View article: An Early Adverse Drug Event Detection Approach with False Discovery Rate Control
An Early Adverse Drug Event Detection Approach with False Discovery Rate Control Open
Adverse drug event (ADE) is a significant challenge in clinical practice. Many ADEs have not been identified timely after the approval of the corresponding drugs. Despite the use of drug similarity network demonstrates early success on imp…
View article: Towards MoE Deployment: Mitigating Inefficiencies in Mixture-of-Expert (MoE) Inference
Towards MoE Deployment: Mitigating Inefficiencies in Mixture-of-Expert (MoE) Inference Open
Mixture-of-Experts (MoE) models have gained popularity in achieving state-of-the-art performance in a wide range of tasks in computer vision and natural language processing. They effectively expand the model capacity while incurring a mini…
View article: Efficiently Upgrading Multilingual Machine Translation Models to Support More Languages
Efficiently Upgrading Multilingual Machine Translation Models to Support More Languages Open
With multilingual machine translation (MMT) models continuing to grow in size and number of supported languages, it is natural to reuse and upgrade existing models to save computation as data becomes available in more languages. However, a…
View article: Hybrid Transducer and Attention based Encoder-Decoder Modeling for Speech-to-Text Tasks
Hybrid Transducer and Attention based Encoder-Decoder Modeling for Speech-to-Text Tasks Open
Yun Tang, Anna Sun, Hirofumi Inaguma, Xinyue Chen, Ning Dong, Xutai Ma, Paden Tomasello, Juan Pino. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.
View article: Efficiently Upgrading Multilingual Machine Translation Models to Support More Languages
Efficiently Upgrading Multilingual Machine Translation Models to Support More Languages Open
With multilingual machine translation (MMT) models continuing to grow in size and number of supported languages, it is natural to reuse and upgrade existing models to save computation as data becomes available in more languages. However, a…
View article: Fixing MoE Over-Fitting on Low-Resource Languages in Multilingual Machine Translation
Fixing MoE Over-Fitting on Low-Resource Languages in Multilingual Machine Translation Open
Sparsely gated Mixture of Experts (MoE) models have been shown to be a compute-efficient method to scale model capacity for multilingual machine translation. However, for low-resource tasks, MoE models severely over-fit. We show effective …
View article: Fixing MoE Over-Fitting on Low-Resource Languages in Multilingual Machine Translation
Fixing MoE Over-Fitting on Low-Resource Languages in Multilingual Machine Translation Open
Sparsely gated Mixture of Experts (MoE) models have been shown to be a compute-efficient method to scale model capacity for multilingual machine translation. However, for low-resource tasks, MoE models severely over-fit. We show effective …
View article: What has Belt and Road gas cooperation brought us for —— The financial benefits, the modern governance, or the narrowed distance between cultures?
What has Belt and Road gas cooperation brought us for —— The financial benefits, the modern governance, or the narrowed distance between cultures? Open
China proposed the Belt and Road Initiative (BRI) in 2013, providing itself with a stable foundation to deepen the gas cooperation with signatories. However, such activities would inevitably confront the cross-cultural challenges that go h…
View article: No Language Left Behind: Scaling Human-Centered Machine Translation
No Language Left Behind: Scaling Human-Centered Machine Translation Open
Driven by the goal of eradicating language barriers on a global scale, machine translation has solidified itself as a key focus of artificial intelligence research today. However, such efforts have coalesced around a small subset of langua…
View article: stopes - Modular Machine Translation Pipelines
stopes - Modular Machine Translation Pipelines Open
Pierre Andrews, Guillaume Wenzek, Kevin Heffernan, Onur Çelebi, Anna Sun, Ammar Kamran, Yingzhe Guo, Alexandre Mourachko, Holger Schwenk, Angela Fan. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: S…
View article: Turing-Hopf bifurcation analysis in a diffusive Gierer-Meinhardt model
Turing-Hopf bifurcation analysis in a diffusive Gierer-Meinhardt model Open
The reaction-diffusion Gierer-Meinhardt system in one dimensional bounded domain is considered in the present paper. The Hopf bifurcation is investigated, which is found to be degenerate. With the aid of Maple, the normal form associated w…
View article: Confucianism and Daoism: From Max Weber to the Present
Confucianism and Daoism: From Max Weber to the Present Open
It has been over a century since the 1915 publication of Max Weber's The Religion of China: Confucianism and Daoism.In the past decades, we have been wrestling with Weber not only over the issue of the rise of the spirit of modern capitali…