[Theory] REMINDER: 7/8 Thesis Defense: Mingda Chen, TTIC

Mary Marre mmarre at ttic.edu
Fri Jul 1 18:39:55 CDT 2022


*Thesis Defense: Mingda Chen, TTIC*

*When:  *     Friday*,* July 8th at *12:00 - 2:00 pm CT*



*Virtually: *  *attend virtually here
<https://uchicago.zoom.us/j/98474647709?pwd=SnFHdzh0VFkvT1k2UndLdkJMTmpadz09>*



*Who: *        Mingda Chen, TTIC


*Thesis title: *Leveraging Natural Supervision for Language Representation
Learning and Generation

*Abstract: *Recent breakthroughs in Natural Language Processing (NLP) have
been driven by language models trained on a massive amount of plain text.
While powerful, deriving supervision from textual resources is still an
open question. For example, language model pretraining often neglects the
rich, freely-available structures in textual data. In this thesis, we
describe three lines of work that seek to improve the training and
evaluation of neural models using naturally-occurring supervision.

We first investigate self-supervised training losses to help enhance the
performance of pretrained language models for various NLP tasks.
Specifically, we alter the sentence prediction loss to make it better
suited to other pretraining losses and more challenging to solve. We design
an intermediate finetuning step that uses self-supervised training to
promote models' ability in cross-task generalization.

Then we describe methods to leverage the structures in Wikipedia and
paraphrases. In particular, we propose training losses to exploit
hyperlinks, article structures, and article category graphs for entity-,
discourse-, entailment-related knowledge. We propose a framework that uses
paraphrase pairs to disentangle semantics and syntax in sentence
representations. We extend the framework for a novel generation task that
controls the syntax of output text with a sentential exemplar.

Lastly, we discuss our work on tailoring textual resources for establishing
challenging evaluation tasks. We introduce three datasets by defining novel
tasks using various fan-contributed websites, including a long-form
data-to-text generation dataset, a screenplay summarization dataset, and a
long-form story generation dataset. These datasets have unique
characteristics offering challenges to future work in their respective task
settings.

*Thesis committee:* Karen Livescu, Sam Wiseman, Luke Zettlemoyer

*Thesis Advisor:* *Kevin Gimpel* <kgimpel at ttic.edu>




Mary C. Marre
Faculty Administrative Support
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Chicago, IL  60637*
*mmarre at ttic.edu <mmarre at ttic.edu>*


On Fri, Jun 24, 2022 at 4:43 PM Mary Marre <mmarre at ttic.edu> wrote:

> *Thesis Defense: Mingda Chen, TTIC*
>
> *When:  *     Friday*,* July 8th at *12:00 - 2:00 pm CT*
>
>
>
> *Virtually: *  *attend virtually here
> <https://uchicago.zoom.us/j/98474647709?pwd=SnFHdzh0VFkvT1k2UndLdkJMTmpadz09>*
>
>
>
> *Who: *        Mingda Chen, TTIC
>
>
> *Thesis title: *Leveraging Natural Supervision for Language
> Representation Learning and Generation
>
> *Abstract: *Recent breakthroughs in Natural Language Processing (NLP)
> have been driven by language models trained on a massive amount of plain
> text. While powerful, deriving supervision from textual resources is still
> an open question. For example, language model pretraining often neglects
> the rich, freely-available structures in textual data. In this thesis, we
> describe three lines of work that seek to improve the training and
> evaluation of neural models using naturally-occurring supervision.
>
> We first investigate self-supervised training losses to help enhance the
> performance of pretrained language models for various NLP tasks.
> Specifically, we alter the sentence prediction loss to make it better
> suited to other pretraining losses and more challenging to solve. We design
> an intermediate finetuning step that uses self-supervised training to
> promote models' ability in cross-task generalization.
>
> Then we describe methods to leverage the structures in Wikipedia and
> paraphrases. In particular, we propose training losses to exploit
> hyperlinks, article structures, and article category graphs for entity-,
> discourse-, entailment-related knowledge. We propose a framework that uses
> paraphrase pairs to disentangle semantics and syntax in sentence
> representations. We extend the framework for a novel generation task that
> controls the syntax of output text with a sentential exemplar.
>
> Lastly, we discuss our work on tailoring textual resources for
> establishing challenging evaluation tasks. We introduce three datasets by
> defining novel tasks using various fan-contributed websites, including a
> long-form data-to-text generation dataset, a screenplay summarization
> dataset, and a long-form story generation dataset. These datasets have
> unique characteristics offering challenges to future work in their
> respective task settings.
>
> *Thesis committee:* Karen Livescu, Sam Wiseman, Luke Zettlemoyer
>
> *Thesis Advisor:* *Kevin Gimpel* <kgimpel at ttic.edu>
>
>
>
>
> Mary C. Marre
> Faculty Administrative Support
> *Toyota Technological Institute*
> *6045 S. Kenwood Avenue*
> *Chicago, IL  60637*
> *mmarre at ttic.edu <mmarre at ttic.edu>*
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/theory/attachments/20220701/f7f66e84/attachment.html>


More information about the Theory mailing list