Title: Building and Evaluating Controllable Models for Text Simplification
Date: Friday, December 16, 2022
Time: 1pm - 3pm EST
Location: Coda C0903 Ansley & Zoom link
Mounica Maddela
PhD Student in Computer Science
School of Interactive Computing
College of Computing
Georgia Institute of Technology
Committee
Dr. Wei Xu (Advisor), School of Interactive Computing, Georgia Tech
Dr. Alan Ritter, School of Interactive Computing, Georgia Tech
Dr. Mark Riedl, School of Interactive Computing, Georgia Tech
Dr. Colin Cherry, Google Research
Dr. Y-Lan Boureau, Meta AI Research
Abstract
Although the existing natural language generation systems (NLG) have made great progress in generating fluent text indistinguishable from human written text, they still lack the capability to adapt to specific constraints or attributes in practical applications. There has been an emerging trend in NLG to develop controllable methods for text generation that generate texts by controlling different attributes such as sentiment, formality, politeness, and topic.
In this thesis, I focus on controllable text generation for the task of Automatic Text Simplification (ATS). ATS aims to improve the readability of texts with simpler grammar and word choices while preserving the original meaning. It is an audience-dependent task because the readability constraints vary based on the target population. Therefore, controllability is essential for the ATS systems to generate text adhering to diverse constraints. An ideal automatic simplification system should be able to control various attributes of the generated text such as syntactic structures, length, readability levels, and word choices that are appropriate for the situation. However, the existing simplification systems lack the capability to adapt to different readability constraints.
In this work, I develop novel controllable approaches for ATS that combine linguistic rules with neural approaches to generate simplified text at different readability levels. Apart from building controllable systems, I also propose reliable human evaluation and automatic evaluation approaches to assess machine-generated simplified text.