[Theory] TODAY: 2/15 Talks at TTIC: Jungo Kasai, University of Washington

[Mary Marre]

*When:*        Wednesday, February 15, 2023 at* 11:30** a**m CT   *


*Where:       *Talk will be given *live, in-person* at

                   TTIC, 6045 S. Kenwood Avenue

                   5th Floor, Room 530


*Virtually:*  *via* Panopto (*livestream
<https://uchicago.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=ca450d5a-9033-4f96-8dce-afa40007dbcf>*
)


*Who: *        Jungo Kasai, University of Washington


------------------------------
*Title: *Democratized Natural Language Processing: Efficiency, Evaluation,
and Beyond

*Abstract: *In this talk, I advocate for the democratization of natural
language processing (NLP) knowledge, where people with diverse backgrounds
and groups with diverse budgets can: build, use, analyze, and evaluate
models; collaborate to solve research problems; and accelerate advances in
NLP. I present three key research contributions towards realizing this
long-term goal. First, I will show an approach to improving the efficiency
and accessibility of widely-available transformer models, while retaining
their performance. Second, I discuss collaborative evaluation frameworks
that facilitate reliable evaluations using diverse metrics. Third, I
describe a language generation algorithm that customizes general-purpose
models for specialized, low-resource domains. I will then conclude by
highlighting future directions towards dynamic and massively multilingual
applications that are needed in the real world.

*Bio: *Jungo Kasai is a fifth-year Ph.D. student in the Paul G. Allen
School of Computer Science & Engineering at the University of Washington,
advised by Noah A. Smith. His research focuses on efficient and accessible
models, evaluation methodologies, and language generation algorithms for
natural language processing and artificial intelligence. He is supported by
an IBM Ph.D. fellowship, the Masason foundation, and a Funai overseas
scholarship. His papers have been accepted to top-tier conferences, such as
ACL, NAACL, ICML, and EMNLP, receiving a best paper award at NAACL 2022.

*Host:* Karen Livescu <klivescu at ttic.edu>

Mary C. Marre
Faculty Administrative Support
*Toyota Technological Institute*
*6045 S. Kenwood Avenue, Rm 517*
*Chicago, IL  60637*
*773-834-1757*
*mmarre at ttic.edu <mmarre at ttic.edu>*


On Tue, Feb 14, 2023 at 2:59 PM Mary Marre <mmarre at ttic.edu> wrote:

> *When:*        Wednesday, February 15, 2023 at* 11:30** a**m CT   *
>
>
> *Where:       *Talk will be given *live, in-person* at
>
>                    TTIC, 6045 S. Kenwood Avenue
>
>                    5th Floor, Room 530
>
>
> *Virtually:*  *via* Panopto (*livestream
> <https://uchicago.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=ca450d5a-9033-4f96-8dce-afa40007dbcf>*
> )
>
>
> *Who: *        Jungo Kasai, University of Washington
>
>
> ------------------------------
> *Title: *Democratized Natural Language Processing: Efficiency,
> Evaluation, and Beyond
>
> *Abstract: *In this talk, I advocate for the democratization of natural
> language processing (NLP) knowledge, where people with diverse backgrounds
> and groups with diverse budgets can: build, use, analyze, and evaluate
> models; collaborate to solve research problems; and accelerate advances in
> NLP. I present three key research contributions towards realizing this
> long-term goal. First, I will show an approach to improving the efficiency
> and accessibility of widely-available transformer models, while retaining
> their performance. Second, I discuss collaborative evaluation frameworks
> that facilitate reliable evaluations using diverse metrics. Third, I
> describe a language generation algorithm that customizes general-purpose
> models for specialized, low-resource domains. I will then conclude by
> highlighting future directions towards dynamic and massively multilingual
> applications that are needed in the real world.
>
> *Bio: *Jungo Kasai is a fifth-year Ph.D. student in the Paul G. Allen
> School of Computer Science & Engineering at the University of Washington,
> advised by Noah A. Smith. His research focuses on efficient and accessible
> models, evaluation methodologies, and language generation algorithms for
> natural language processing and artificial intelligence. He is supported by
> an IBM Ph.D. fellowship, the Masason foundation, and a Funai overseas
> scholarship. His papers have been accepted to top-tier conferences, such as
> ACL, NAACL, ICML, and EMNLP, receiving a best paper award at NAACL 2022.
>
> *Host:* Karen Livescu <klivescu at ttic.edu>
>
>
> Mary C. Marre
> Faculty Administrative Support
> *Toyota Technological Institute*
> *6045 S. Kenwood Avenue, Rm 517*
> *Chicago, IL  60637*
> *773-834-1757*
> *mmarre at ttic.edu <mmarre at ttic.edu>*
>
>
> On Wed, Feb 8, 2023 at 7:02 PM Mary Marre <mmarre at ttic.edu> wrote:
>
>> *When:*        Wednesday, February 15, 2023 at* 11:30** a**m CT   *
>>
>>
>> *Where:       *Talk will be given *live, in-person* at
>>
>>                    TTIC, 6045 S. Kenwood Avenue
>>
>>                    5th Floor, Room 530
>>
>>
>> *Virtually:*  *via* Panopto (*livestream
>> <https://uchicago.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=ca450d5a-9033-4f96-8dce-afa40007dbcf>*
>> )
>>
>>
>> *Who: *        Jungo Kasai, University of Washington
>>
>>
>> ------------------------------
>> *Title: *Democratized Natural Language Processing: Efficiency,
>> Evaluation, and Beyond
>>
>> *Abstract: *In this talk, I advocate for the democratization of natural
>> language processing (NLP) knowledge, where people with diverse backgrounds
>> and groups with diverse budgets can: build, use, analyze, and evaluate
>> models; collaborate to solve research problems; and accelerate advances in
>> NLP. I present three key research contributions towards realizing this
>> long-term goal. First, I will show an approach to improving the efficiency
>> and accessibility of widely-available transformer models, while retaining
>> their performance. Second, I discuss collaborative evaluation frameworks
>> that facilitate reliable evaluations using diverse metrics. Third, I
>> describe a language generation algorithm that customizes general-purpose
>> models for specialized, low-resource domains. I will then conclude by
>> highlighting future directions towards dynamic and massively multilingual
>> applications that are needed in the real world.
>>
>> *Bio: *Jungo Kasai is a fifth-year Ph.D. student in the Paul G. Allen
>> School of Computer Science & Engineering at the University of Washington,
>> advised by Noah A. Smith. His research focuses on efficient and accessible
>> models, evaluation methodologies, and language generation algorithms for
>> natural language processing and artificial intelligence. He is supported by
>> an IBM Ph.D. fellowship, the Masason foundation, and a Funai overseas
>> scholarship. His papers have been accepted to top-tier conferences, such as
>> ACL, NAACL, ICML, and EMNLP, receiving a best paper award at NAACL 2022.
>>
>> *Host:* Karen Livescu <klivescu at ttic.edu>
>>
>>
>> Mary C. Marre
>> Faculty Administrative Support
>> *Toyota Technological Institute*
>> *6045 S. Kenwood Avenue, Rm 517*
>> *Chicago, IL  60637*
>> *773-834-1757*
>> *mmarre at ttic.edu <mmarre at ttic.edu>*
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/theory/attachments/20230215/4680091f/attachment-0001.html>