Ting-Hao (Kenneth) Huang 黃挺豪

Tenure-Track Assistant Professor
College of Information Sciences and Technology (IST)
Pennsylvania State University (University Park)

We are hiring! Come work with us!

I combine AI with crowdsourcing to create systems that are more usable, robust, and intelligent.

Office: E357 Westgate Building
Office Hour: 1pm-2pm Mon (Spring 2019)
Email: txh710@psu.edu

News

The pre-print of ACL 2019 paper, Visual Story Post-Editing, is out!
Our short paper is accepted by ACL 2019!
The paper, "Smell Pittsburgh: Community-Empowered Mobile Smell Reporting System", won the Outstanding Paper Award in IUI 2019!
Our demo paper, "Dixit: Interactive Visual Storytelling via Term Manipulation", is accepted by TheWebConf 2019.
Our extended abstract, "On How Users Edit Computer-Generated Visual Stories", is accepted by #CHI2019 Late-Breaking-Work. This is the FIRST PAPER from Crowd-AI Lab!

Teaching

Crowdsourcing & Crowd-AI Systems (IST 597, Spring 2019)

Selected Publications

Evorus is a crowd-powered conversational assistant built to automate itself over time. Our 5-month-long deployment shows that Evorus can automate itself without compromising conversation quality.

Evorus: A Crowd-Powered Conversational Assistant Built to Automate Itself Over Time

Ting-Hao K. Huang, Joseph Chee Chang, Jeffrey P. Bigham.

In Proceedings of Conference on Human Factors in Computing Systems 2018 (CHI 2018), 2018, Montréal, Canada.

Media Coverage: [CMU News] [EurekAlert!] [TribLIVE] [pressetext] [WTAE TV News]

Best Paper Honorable Mention Award (101/2500 = 5%)

[Paper] [arXiv] [Video]
Retainer is fast but costly, posting HIT is cheap but slow. Deployed Chorus uses Ignition to balance them.

A 10-Month-Long Deployment Study of On-Demand Recruiting for Low-Latency Crowdsourcing

Ting-Hao K. Huang, Jeffrey P. Bigham.

In Proceedings of The fifth AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2017), 2017, Quebec City, Canada.

[Paper] [Official] [Slides] [Video Demo]
Chorus was launched to public in May, 2016. The system details are described in our HCOMP 2016 paper.

"Is there anything else I can help you with?": Challenges in Deploying an On-Demand Crowd-Powered Conversational Agent

Ting-Hao K. Huang, Walter S. Lasecki, Amos Azaria, Jeffrey P. Bigham.

In Proceedings of Conference on Human Computation & Crowdsourcing (HCOMP 2016), 2016, Austin, TX, USA.

Media Coverage: [The Register]

[Paper] [arXiv] [Slides]
We developed "Guardian" to convert an existent Web APIs to a dialog system by using crowdsourcing.

Guardian: A Crowd-Powered Spoken Dialog System for Web APIs

Ting-Hao K. Huang, Walter S Lasecki, Jeffrey P Bigham.

In Proceedings of Conference on Human Computation & Crowdsourcing (HCOMP 2015), pages 62–71, November, 2015, San Diego, USA.

[Paper] [Official] [BIB] [Slides]
A group of crowd workers can reliably extract target entities on demand from a short dialog in an average of ~8 seconds.

Real-time On-Demand Crowd-powered Entity Extraction

Ting-Hao K. Huang, Yun-Nung Chen, Jeffrey P. Bigham.

In Proceedings of the 5th Edition Of The Collective Intelligence Conference (CI 2017, oral presentation), 2017, New York University, NY, USA.

[Paper] [arXiv] [Data & Code] [Interface Demo]
I developed the Visual Storytelling Dataset during my summer internship at Microsoft Research in 2015.

Visual Storytelling

Ting-Hao K. Huang*, Francis Ferraro*, Nasrin Mostafazadeh, Ishan Misra, Jacob Devlin, Aishwarya Agrawal, Ross Girshick, Xiaodong He, Pushmeet Kohli, Dhruv Batra, Larry Zitnick, Devi Parikh, Lucy Vanderwende, Michel Galley and Margaret Mitchell. (*: Co-first author.)

In Proc. NAACL 2016, June, 2016, San Diego, CA, USA.

Media Coverage: [MIT Technology Review] [The Stack] [Microsoft Official Blog] [Venture Beat] [WIRED UK]

[Paper] [Poster] [Website]

Honors

Best Paper Honorable Mention Award (101/2500 = 5%), CHI 2018.
Best Paper Honorable Mention Award (14/281 = 5%), CHI LBW 2016.
Yahoo! Fellowship of the InMind project at CMU, 2014 – Present.
Best Poster Award, LTI Student Research Symposium 2013.