<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
<br class="">
<div class=""><br class="Apple-interchange-newline">
</div>
<div dir="auto" class="" style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;">
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
This is an announcement of Qiming Wang’s Candidacy Exam.</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
===============================================</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Candidate: Qiming Wang<span class="apple-converted-space"> </span></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Date: Thursday, September 21st,2023</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Time: 2pm CST</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Location: JCL 236</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Title: Data Discovery using Natural Language Questions via a Self-Supervised Approach</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Abstract: Most deployed data discovery systems, such as Google Datasets, and open data portals only support keyword search. Keyword search is geared towards general audiences but limits the types of queries the systems can answer. We propose a new system that
lets users write natural language questions directly. A major barrier to using this learned data discovery system is that it needs expensive-to-collect training data, thus limiting its utility.<br class="">
In this paper, we introduce a self-supervised approach to assemble training datasets and train learned discovery systems without human intervention. It requires addressing several challenges, including the design of self-supervised strategies for data discovery,
table representation strategies to feed to the models, and relevance models that work well with the synthetically generated questions. We combine all the above contributions into a system, Solo, that solves the problem end to end. The evaluation results demonstrate
the new techniques outperform state-of-the-art approaches on well-known benchmarks. All in all, the technique is a stepping stone towards building learned discovery systems.</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Advisors: Raul Castro Fernandez</div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
<o:p class=""> </o:p></div>
<div class="" style="margin: 0in; font-size: 10pt; font-family: "Helvetica Neue";">
Committee Members: Raul Castro Fernandez, Chard Kyle, Sanjay Krishnan</div>
<div class=""><br class="">
</div>
</div>
</body>
</html>