<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">
<div class="">This is an announcement of Xiaoan Ding's Dissertation Defense.</div>
<div class="">===============================================</div>
<div class="">Candidate: Xiaoan Ding</div>
<div class=""><br class="">
</div>
<div class="">Date: Monday, June 13, 2022</div>
<div class=""><br class="">
</div>
<div class="">Time: 1 pm CST</div>
<div class=""><br class="">
</div>
<div class="">Remote Location: <a href="https://uchicago.zoom.us/j/92085734782?pwd=SlhOek9ZdW1TcVl0YmRkcGh3ZkFDUT09" class="">https://uchicago.zoom.us/j/92085734782?pwd=SlhOek9ZdW1TcVl0YmRkcGh3ZkFDUT09</a></div>
<div class=""><br class="">
</div>
<div class=""><br class="">
</div>
<div class="">Title: BEYOND ACCURACY: MODELING TEXT FOR ROBUST NLP</div>
<div class=""><br class="">
</div>
<div class="">Abstract: Research in model robustness has a long history. Improving model robustness generally refers to the goal of ensuring machine learning models are resistant to a variety of imperfect training and testing conditions. With the unprecedented
progress in deep learning architectures, large-scale training, and learning algorithms, pretrained models have become pivotal in AI. However, when considering real-world scenarios, these models are still fragile and brittle, which impedes the safe deployment
of NLP models in production systems.</div>
<div class=""><br class="">
</div>
<div class="">In this work, we consider wide applications in NLP and define model robustness to broader aspects: (1) data-efficient: models can adapt to new domains with limited annotated data in both pretrained-finetuned and trained-from-scratch set-ups; (2)
resilient: models can perform reliably under uncertainties and challenging circumstances; (3) fair: predictors or generators can make safe decisions and filter undesirable biases especially those imbued with toxicity, hate, and social bias; (4) trusted: models
are expected to yield factual and faithful content. To tackle these robustness issues, on the modeling side, we explored both discrete and continuous latent-variable generative models and various graphical model configurations; on the learning algorithms side,
we investigated generative pretraining and various discriminative finetuning objectives in generative classifiers, gradient-based optimization and importance-sampled log marginal likelihood on learning deep latent-variable models; on the applications side,
we developed document classifiers, textual relation predictors, a controllable story generator, and a hallucinated content detector.</div>
<div class=""><br class="">
</div>
<div class="">Advisors: Kevin Gimpel and Janos Simon</div>
<div class=""><br class="">
</div>
<div class="">Committee Members: Kevin Gimpel, Janos Simon, Chenhao Tan, and Samuel Wiseman</div>
<div class=""><br class="">
</div>
<div class=""> <a href="https://drive.google.com/file/d/1N_V5H-QmjcUX5f13Rda15Mvyx813FF51/view?usp=sharing" class="">https://drive.google.com/file/d/1N_V5H-QmjcUX5f13Rda15Mvyx813FF51/view?usp=sharing</a></div>
</body>
</html>