avatar

Qingqing Cao | 曹庆庆

Postdoc at UW


News

  • 2023-05 Two papers accepted to ACL 2023

  • 2023-04 Checkout our survey paper on efficient NLP (accepted to TACL)

  • 2022-04 MobiVQA paper accepted to IMWUT/UbiComp 2022

  • 2021-10

    I started my next journey as a postdoc at UW

  • 2021-09

    I successfully defended!

  • 2021-06

    Our energy estimation paper (IrEne) has been chosen as an oral presentation at ACL 2021!

  • 2021-06

    Mobile DNN accelerator paper accepted to the EMDL 2021 workshop @ MobiSys!

  • 2021-05

    One long paper accepted to ACL 2021

  • 2021-03

    Thrilled to receive the Catacosinos Fellowship!

  • 2021-02

    Present my research (poster) on Stony Brook CS Graduate Research Day.

  • 2021-01

    Attended UChicago Rising Stars in Data Science Workshop, check out my 5-min lightning talk

  • 2020-12

    Joined the JSys Student Editorial Board

  • 2020-12

    Excited to be a CDAC Rising Star

  • 2020-10

    Passed the thesis proposal defense! Feel free to check out the slides.

  • 2020-09

    One paper accepted to SustaiNLP workshop @ EMNLP 2020

  • 2020-09

    Happy to serve on the EuroSys 2021 shadow PC

  • 2020-07

    Attended amazing ACL 2020!

  • 2020-04

    One long paper accepted to ACL 2020

  • 2020-02

    Invited to serve on the PC for ACL 2020 (demo track)

  • 2019-06

    Attending MobiSys 2019

  • 2019-03

    DeQA accepted to MobiSys 2019

  • 2018-05

    Starting Intership at Microsoft Research Redmond

  • 2018-03

    Serving on MobiSys 2018 PhD Forum

  • 2017-10

    Attending MobiCom 2017

  • 2017-07

    Starting Internship at Bell-Labs Cambridge

  • 2017-06

    UIWear accepted at MobiCom 2017

  • 2017-05

    Awarded MobiSys 2017 Student Travel Grant

  • 2017-05

    MobiRNN accepted at EMDL 2017

Bio

Bio

Qingqing Cao is a postdoc working with Prof. Hannaneh Hajishirzi in the UW NLP group at the University of Washington. He obtained a computer science Ph.D. at Stony Brook University where he worked with Prof. Aruna Balasubramanian and Prof. Niranjan Balasubramanian. His research interests include efficient NLP, mobile computing, and ML systems. He has focused on building efficient and practical NLP systems for both edge devices and the cloud, such as on-device (visual) question answering (MobiSys 2019, IMWUT 2022), faster Transformer models (ACL 2020), and accurate energy estimation of NLP models (ACL 2021). He was a recipient of the Catacosinos Fellowship at Stony Brook University and a Rising Star in Data Science at the University of Chicago. His website is awk.ai.
Info

Info

Note: If you are an undergraduate or master’s student and want to work on research projects with me, send me an email with the subject “student research application”.
Other research topics include: (1) improving storage/memory/compute efficiency of large language models (LLMs) and retrieval/generation models; (2) building fast/light-weight/on-device language and vision systems; I am generally interested in building efficient models and practical systems for sustainable NLP/AI, so I am happy to collaborate on other possible projects in efficient AI areas.
Education

Education

  • Ph.D. in Computer Science,  Stony Brook University

    Aug 2015 - Sep 2021

  • B.Eng. in Computer Science and Technology,  Wuhan University

    Sept 2011 - Jun 2015

Experience

Experience

  • Postdoctoral Scholar,  University of Washington

    Oct 2021 - present

  • Research Assistant,  Stony Brook University

    June, 2016 - Sep 2021

  • Research Intern,  Microsoft Research Redmond

    June, 2018 - Aug, 2018

    Mentor: Oriana Riva

  • Research Intern,  Bell Labs Cambridge

    July, 2017 - Sept, 2017

    Mentor: Nic Lane

Publications

Publications

NeurIPS 2023Adaptive Representations for Semantic Search
Aniket Rege*, Aditya Kusupati*, Sharan Ranjit, Alan Fan, Qingqing Cao, Sham Kakade, Prateek Jain, and Ali Farhadi.
Neural Information Processing Systems
arxiv, code

preprintEfficiency Pentathlon: A Standardized Arena for Efficiency Evaluation
Hao Peng, Qingqing Cao, Jesse Dodge, Matthew E. Peters, Jared Fernandez, Tom Sherborne, Kyle Lo, Sam Skjonsberg, Emma Strubell, Darrell Plessas, Iz Beltagy, Evan Pete Walsh, Noah A. Smith, Hannaneh Hajishirzi.
Under submission
arxiv, code

ACL 2023 (long paper)PuMer: Pruning and Merging Tokens for Efficient Vision Language Models
Qingqing Cao, Bhargavi Paranjape and Hannaneh Hajishirzi.
The 61th Annual Meeting of the Association for Computational Linguistics.
paper, arxiv, acl, code, poster, slides

ACL 2023 (long paper)A Survey for Efficient Open Domain Question Answering
Qin Zhang, Shangsi Chen, Dongkuan Xu, Qingqing Cao, Xiaojun Chen, Trevor Cohn and Meng Fang.
The 61th Annual Meeting of the Association for Computational Linguistics.
arxiv, acl

TACL 2023Efficient Methods for Natural Language Processing: A Survey
Marcos Treviso, Ji-Ung Lee, Tianchu Ji, Betty van Aken, Qingqing Cao, Manuel R. Ciosici, Michael Hassid, Kenneth Heafield, Sara Hooker, Colin Raffel, Pedro H. Martins, André F. T. Martins, Jessica Zosa Forde, Peter Milder, Edwin Simpson, Noam Slonim, Jesse Dodge, Emma Strubell, Niranjan Balasubramanian, Leon Derczynski, Iryna Gurevych, Roy Schwartz.
Transactions of the Association for Computational Linguistics.
paper

IMWUT/UbiComp 2022MobiVQA: Efficient On-Device Visual Question Answering
Qingqing Cao, Prerna Khanna, Aruna Balasubramanian and Nicholas D. Lane.
Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.
pdf, bibtex, slides

EMNLP 2021 (demo)IrEne-viz: Visualizing Energy Consumption of Transformer Models
Yash Kumar Lal, Reetu Singh, Harsh Trivedi, Qingqing Cao, Aruna Balasubramanian, Niranjan Balasubramanian.
The 2021 Conference on Empirical Methods in Natural Language Processing.
pdf, demo url, bibtex, acl

ACL 2021 (long paper)IrEne: Interpretable Energy Prediction for Transformers
Qingqing Cao, Yash Kumar Lal, Harsh Trivedi, Aruna Balasubramanian and Niranjan Balasubramanian.
The 59th Annual Meeting of the Association for Computational Linguistics.
pdf, code, bibtex, acl, gslides

EMDL@MobiSys 2021Are Mobile DNN Accelerators Accelerating DNNs?
Qingqing Cao, Alexandru Eugen Irimiea, Mohamed Abdelfattah, Aruna Balasubramanian and Nicholas D. Lane.
The 5th International Workshop on Embedded and Mobile Deep Learning.
pdf, slides, code, bibtex, gslides, presentation

SustaiNLP@EMNLP 2020Towards Accurate and Reliable Energy Measurement of NLP Models
Qingqing Cao, Aruna Balasubramanian and Niranjan Balasubramanian.
The First Workshop on Simple and Efficient Natural Language Processing.
pdf, slides, code, bibtex, acl, gslides, presentation

ACL 2020 (long paper)DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering
Qingqing Cao, Harsh Trivedi, Aruna Balasubramanian and Niranjan Balasubramanian.
The 58th Annual Meeting of the Association for Computational Linguistics.
pdf, slides, code, bibtex, acl, gslides, presentation

MobiSys 2019DeQA: On-Device Question Answering
Qingqing Cao, Noah Weber, Niranjan Balasubramanian, and Aruna Balasubramanian.
The 17th ACM International Conference on Mobile Systems, Applications, and Services.
pdf, slides, code, bibtex

MobiCom 2017UIWear: Easily Adapting User Interfaces for Wearable Devices
Jian Xu (co-primary), Qingqing Cao (co-primary), Aditya Prakash, Aruna Balasubramanian, and Don Porter.
The 23rd Annual International Conference on Mobile Computing and Networking.
pdf, slides, code, bibtex

MobiCom 2017Demo: UIWear: Easily Adapting User Interfaces for Wearable Devices
Jian Xu (co-primary), Qingqing Cao (co-primary), Aditya Prakash, Aruna Balasubramanian, and Don Porter.
The 23rd Annual International Conference on Mobile Computing and Networking.
pdf, poster, video, bibtex

EMDL@MobiSys 2017MobiRNN: Efficient Recurrent Neural Network Execution on Mobile GPU
Qingqing Cao, Niranjan Balasubramanian, Aruna Balasubramanian.
The 1st International Workshop on Embedded and Mobile Deep Learning.
pdf, slides, code, bibtex

Talks

Talks

Efficient NLP for Heterogeneous Platforms: Microsoft Research Asia (May 2021), University of Washington (April 2021), UC Santa Barbara (March 2021), University of Glasgow (March 2021)

Answering Questions in a Faster and Greener Way: SBU 3MT (April 2021), talk video

Efficient Neural Question Answering for Heterogeneous Platforms: Uchicago CDAC (Jan 2021), talk video

Honors & Awards

Honors & Awards

Postdoc Research Award, University of Washington, 2022

Catacosinos Fellowship (2 out of 232 PhD students), Stony Brook University, 2021

CDAC Rising Stars in Data Science, University of Chicago, 2021

MobiSys Student Travel Grant, ACM SIGMOBILE, 2017

Special CS Department Chair Fellowship, Stony Brook University, 2015

National Scholarship (top 0.2%), China Ministry of Education, 2013

Service

Service

Program committee/Reviewer: ACL Rolling Review, EMNLP 2021-2022, ACL 2021-2023, NAACL 2021-2022, Eurosys 2021 (shadow), ACL 2020 (demo), MobiSys 2018 (PhD forum), IEEE Transactions on Mobile Computing (reviewer), JSys Student Editorial Board Member.

Volunteering Service: Student volunteer for MobiSys 2017 and ACL 2020.

Mentor service: Stony Brook CS Grad Buddies Program; MS students: Aditya Prakash, Sruti Kumari, Mohit Marwari, Naga Naravamakula, Chenghao Yang, and Alexandru Eugen Irimiea (Oxford)

Teaching: Instructor for Women in Science & Engineering (WISE) 380