TAESIK GONG

Taesik Gong

AI Research Scientist

Bell Labs &
University of Cambridge


I am a Research Scientist at Nokia Bell Labs and a Visiting Researcher at the University of Cambridge. My research aims to provide software solutions that foster practical, adaptive, and efficient on-device AI systems. I earned my Ph.D. and M.S. from School of Computing at KAIST under the supervision of Prof. Sung-Ju Lee, with close collaboration with Prof. Jinwoo Shin. I am a recipient of Google Ph.D. Fellowship in Machine Learning. My Ph.D. dissertation won the Best Ph.D. Dissertation Award from both School of Computing & College of Engineering at KAIST. During my Ph.D., I was a research intern at Google Research/Google AI (2022), Microsoft Research (2019), and Nokia Bell Labs (2018). I received my B.S. in Computer Science from Yonsei University with Summa Cum Laude.

News

  • Mar. 2024 | ๐Ÿ‘” Visiting Researcher at the University of Cambridge, Cambridge, UK
  • Feb. 2024 | ๐Ÿ“ Paper accepted at CVPR โ€˜24
  • Jan. 2024 | ๐Ÿ—ฃ๏ธ Session Speaker at HSN โ€˜24
  • Dec. 2023 | ๐Ÿ›ซ NeurIPS โ€˜23 @ New Orleans, LA (Dec. 10 - Dec. 15)
  • Oct. 2023 | ๐Ÿ“ Paper accepted at ACM UbiComp โ€˜24
  • Oct. 2023 | ๐Ÿ—ฃ๏ธ Special Session Speaker at ICTC โ€˜23
  • Oct. 2023 | ๐Ÿ›ซ UbiComp โ€˜23 @ Cancun, Mexico (Oct. 8 - 12)
  • Sep. 2023 | ๐Ÿ“ Paper accepted at NeurIPS โ€˜23
  • Aug. 2023 | ๐Ÿ‘” Artifact Evaluation Committee at ACM MobiCom โ€˜23
  • Jun. 2023 | ๐Ÿ‘” Full-time Research Scientist at Nokia Bell Labs, Cambridge, UK
  • May. 2023 | ๐Ÿ“ Paper accepted at INTERSPEECH โ€˜23
  • Apr. 2023 | ๐Ÿ“ Paper accepted at ACM UbiComp โ€˜23
  • Feb. 2023 | ๐Ÿ† Best Ph.D. Dissertation Award from KAIST College of Engineering
  • Feb. 2023 | ๐Ÿ† Best Ph.D. Dissertation Award from KAIST School of Computing
  • Jan. 2023 | ๐Ÿ† AI/CS/EE Rising Stars

Work Experience

University of Cambridge Logo

2024.03.01 - Present
Visiting Researcher, University of Cambridge, Cambridge, UK
(Host: Prof. Cecilia Mascolo)

Nokia Bell Labs Logo

2023.06.01 - Present
Research Scientist, Nokia Bell Labs, Cambridge, UK
(Manager: Prof. Fahim Kawsar)

Google Logo

2022.08.22 - 2022.11.25
Research Intern, Google Research/Google AI, NYC, USA
(Mentors: Josh Belanich, Krishna Somandepalli, Arsha Nagrani, Brian Eoff, Brendan Jou)

Microsoft Logo

2019.04.01 - 2019.06.26
Research Intern, Microsoft Research, Beijing, China
(Mentors: Peng Cheng, Yongqiang Xiong)

Nokia Bell Labs Logo

2018.07.01 - 2018.09.30
Research Intern, Nokia Bell Labs, Cambridge, UK
(Mentors: Sourav Bhattacharya, Alberto Gil Ramos, Akhil Mathur, Fahim Kawsar)

Education

Taesik Gong

2017.08 - 2023.02
Ph.D. in Computer Science, KAIST, Republic of Korea (Advisor: Sung-Ju Lee)
- Dissertation Title: "Overcoming Domain Shift in Mobile Sensing via Machine Learning"
(Best Ph.D. Dissertation Award from both School of Computing & College of Engineering)
- Committee: Sung-Ju Lee, Jinwoo Shin, Sung Ju Hwang, Youngki Lee, Nicholas Lane

Taesik Gong

2016.03 - 2017.08
M.S. in Computer Science, KAIST, Republic of Korea (Advisor: Sung-Ju Lee)

Taesik Gong

2012.03 - 2016.02
B.S. in Computer Science, Yonsei University, Republic of Korea (Summa Cum Laude)


PUBLICATIONS

Preprints

"Collaborative Inference via Dynamic Composition of Tiny AI Accelerators on MCUs"
Taesik Gong*, Si Young Jang*, Utku Gรผnay Acer, Fahim Kawsar, and Chulhong Min
arXiv.
PDF Arxiv

"IMG2IMU: Applying Knowledge from Large-Scale Images to IMU Applications via Contrastive Learning"
Hyungjun Yoon, Hyeongheon Cha, Canh Hoang Nguyen, Taesik Gong, and Sung-Ju Lee
arXiv.
PDF Arxiv

Conferences and Journals

"ATTA: Label-Free Accuracy Estimation for Test-Time Adaptation"
Taeckyung Lee, Sorn Chottananurak, Taesik Gong, Sung-Ju Lee
CVPR 2024

"MIRROR: Towards Generalizable On-Device Video Virtual Try-On for Mobile Shopping"
Dongsig Kang, Eunsu Baek, Sungwook Son, Youngki Lee, Taesik Gong, and Hyung-Sin Kim
UbiComp (IMWUT) 2024
PDF

"SoTTA: Robust Test-Time Adaptation on Noisy Data Streams"
Taesik Gong*, Yewon Kim*, Taeckyung Lee*, Sorn Chottananurak, and Sung-Ju Lee
NeurIPS 2023
PDF Website Arxiv Code

"DAPPER: Label-Free Performance Estimation after Personalization for Heterogeneous Mobile Sensing"
Taesik Gong, Yewon Kim, Adiba Orzikulova, Yunxin Liu, Sung Ju Hwang, Jinwoo Shin, and Sung-Ju Lee
UbiComp (IMWUT) 2023
PDF Website Arxiv

"LanSER: Language-Model Supported Speech Emotion Recognition"
Taesik Gong, Josh Belanich, Krishna Somandepalli, Arsha Nagrani, Brian Eoff, and Brendan Jou
INTERSPEECH 2023
PDF Arxiv Oral Presentation

"Tailoring Interactions: Exploring the Opportune Moment for Remote Computer-mediated Interactions with Home-alone Dogs"
Yewon Kim, Taesik Gong, and Sung-Ju Lee
CHI EA 2023
PDF Website

"MiCrowd: Vision-Based Deep Crowd Counting on MCU"
Sungwook Son, Ahreum Seo, Gyeongseon Eo, Kwangyeon Gill, Taesik Gong, and Hyung-Sin Kim
Sensors 2023
PDF

"NOTE: Robust Continual Test-time Adaptation Against Temporal Correlation"
Taesik Gong, Jongheon Jeong, Taewon Kim, Yewon Kim, Jinwoo Shin, and Sung-Ju Lee
NeurIPS 2022
PDF Website Arxiv Code Qualcomm Innovation Fellowship Finalist

"MyDJ: Sensing Food Intakes with an Attachable on Your Eyeglass Frame"
Jaemin Shin, Seungjoo Lee, Taesik Gong, Hyungjun Yoon, Hyunchul Roh, Andrea Bianchi, and Sung-Ju Lee
CHI 2022
PDF Website Video Honorable Mention

"Adapting to Unknown Conditions in Learning-based Mobile Sensing"
Taesik Gong, Yeonsu Kim, Ryuhaerang Choi, Jinwoo Shin, and Sung-Ju Lee
IEEE Transactions on Mobile Computing (TMC) 2022
PDF Website

"Messaging Beyond Texts with Real-time Image Suggestions"
Joon-Gyum Kim, Taesik Gong, Kyungsik Han, Juho Kim, Jeonggil Ko, and Sung-Ju Lee
MobileHCI 2020
PDF Website

"No More One Liners: Bringing Context into Emoji Recommendations"
Joon-Gyum Kim, Taesik Gong, Bogoan Kim, Jaeyeon Park, Woojeong Kim, Evey Huang, Kyungsik Han, Juho Kim, Jeonggil Ko, and Sung-Ju Lee
ACM Transactions on Social Computing (TSC) 2020
PDF Website

"AudiDoS: Real-Time Denial-of-Service Adversarial Attacks on Deep Audio Models"
Taesik Gong, Alberto Gil C. P. Ramos, Akhil Mathur, Sourav Bhattacharya, and Fahim Kawsar
ICMLA 2019
PDF Oral Presentation

"MetaSense: Few-Shot Adaptation to Untrained Conditions in Deep Mobile Sensing"
Taesik Gong, Yeonsu Kim, Jinwoo Shin, and Sung-Ju Lee
SenSys 2019
PDF Website Video Code

"Knocker: Vibroacoustic-based Object Recognition with Smartphones"
Taesik Gong, Hyunsung Cho, Bowon Lee, and Sung-Ju Lee
UbiComp (IMWUT) 2019
PDF Website Video

"Use MU-MIMO at Your Own Risk โ€“ Why We Don't Get Gb/s Wi-Fi"
Hyunwoo Choi, Taesik Gong, Jaehun Kim, Jaemin Shin, and Sung-Ju Lee
Ad Hoc Networks 2019

PDF Website

"Identifying Everyday Objects with a Smartphone Knock"
Taesik Gong, Hyunsung Cho, Bowon Lee, and Sung-Ju Lee
CHI EA 2018
PDF Website Video

"Enjoy the Silence: Noise Control with Smartphones"
Taesik Gong, Jun Hyuk Chang, Joon-Gyum Kim, Soowon Kang, Donghwi Kim, and Sung-Ju Lee
ICCCN 2017
PDF Website

Posters, Demos, and Videos

"Real-Time Object Identification with a Smartphone Knock"
Taesik Gong, Hyunsung Cho, Bowon Lee, and Sung-Ju Lee
MobiSys 2019
PDF Video Best Video Award

"Towards Condition-Independent Deep Mobile Sensing"
Taesik Gong, Yeonsu Kim, Jinwoo shin, and Sung-Ju Lee
MobiSys 2019
PDF

"Dissecting 802.11ac Performance - Why We Should Turn Off MU-MIMO"
Hyeonwoo Choi, Taesik Gong, Jaehun Kim, Jaemin Shin, and Sung-Ju Lee
MobiSys 2019
PDF

"Bringing Context into Emoji Recommendations"
JoonGyum Kim, Taesik Gong, Bogoan Kim, JaeYeon Park, Woojeong Kim, Evey Huang, Kyungsik Han, Juho Kim, JeongGil Ko, and Sung-Ju Lee
MobiSys 2019
PDF


HONORS

Best Ph.D. Dissertation Award
For the top 10 out of 491 (2%) Ph.D. dissertations, College of Engineering, KAIST, 2023

Best Ph.D. Dissertation Award
For the top 3 out of 41 (7%) Ph.D. dissertations, School of Computing, KAIST, 2023

AI/CS/EE Rising Stars
For early-career researchers who study AI/CS/EE, supported by Google Explore CS Research, 2023

Qualcomm Innovation Fellowship Finalist
For qualified papers in the field of AI/ML, 2022

Google Conference Scholarship
For the primary author with a paper accepted in selected conferences, 2022

Best TA Award
Introduction to Programming (CS101, 2022 Spring), School of Computing, KAIST, 2022

Best Paper Honorable Mention
For the top 5% of all submissions, ACM CHI, 2022

Global Leadership Award
For the most outstanding grad student in creativity, from the President of KAIST, 2021

Google Ph.D. Fellowship
One of the two recipients from the Republic of Korea (only one for Machine Learning), 2020

Best Video Award
ACM MobiSys 2019

NAVER Ph.D. Fellowship
One of the ten Ph.D. students with outstanding research outcomes in KAIST CS Dept., 2018

Outstanding Presentation Award
ACM SIGCHI Korea Local Chapter, 2018

Best Demo/Poster Award
HCI@KAIST Annual Winter Workshop, 2018

Best TA Award
Introduction to Computer Networks (CS341, 2016 Fall), School of Computing, KAIST, 2017


ACADEMIC SERVICES

Committee

  • ACM MobiCom โ€˜23 (Artifact Evaluation Committee)
  • ACM SenSys โ€˜22 (Shadow Program Committee)

Reviewer

  • CVPR โ€˜24
  • ACM IMWUT โ€˜20 โ€˜21 โ€˜22 โ€˜23
  • IEEE TPAMI โ€˜23
  • ACM CHI โ€˜20 โ€˜21 โ€˜24
  • IEEE TMC โ€˜20 โ€˜22
  • IEEE Pervasive Computing โ€˜21
  • ACM TIOT โ€˜23
  • ACM CHI โ€˜20 Case Studies
  • ACM TEI โ€˜20 Work-in-Progress
  • ACM CHI PLAY โ€˜19 Work-in-Progress

PATENTS

โ€œTime-aware Reinforcement-Learning Method for Geographic Ad Hoc Routing with Predefined Mobility and Computer Device for the sameโ€
Jaeha Ahn, Sung-Ju Lee, Taesik Gong, HyungJun Yoon, and Taeckyung Lee
Korea Patent; Filing date: 2022.11.11, No.10-2022-0150662

โ€œReinforcement Learning for Ad-hoc Routing Using Pre-defined Mobilityโ€
Sung-Ju Lee, Taesik Gong, HyungJun Yoon, Taeckyung Lee, and Jaeha Ahn
Korea Patent; Filing date: 2022.03.21, No.10-2022-0034824

โ€œMeta-Learning Based Federated Learning with Small Setโ€
Sung-Ju Lee, Jaewon Jung, Jaemin Shin, and Taesik Gong
Korea Patent; Filing date: 2021.12.13, No.10-2021-0177384

โ€œMethod and Apparatus for Mobile Sensing for Few-Shot Adapting to Untrained Conditionsโ€
Sung-Ju Lee, Taesik Gong, Yeonsu Kim, and Jinwoo Shin
Korea Patent; Filing date: 2019.12.24, No.10-2019-0173544; Issued date: 2023.02.17, No.10-25024410-0000

โ€œMethod and Apparatus for Identifying Object Based on Knockโ€
Sung-Ju Lee, Taesik Gong, Hyunsung Cho, and Bowon Lee
Korea Patent; Filing date: 2018.07.24, No.10-2018-0086161; Issued date: 2019.11.27, No.10-2051493-0000

โ€œActive Noise Control using Smart-phoneโ€
Sung-Ju Lee, Taesik Gong, Jun Hyuk Chang, Joon-Gyum Kim, Soowon Kang, and Donghwi Kim
Korea Patent; Filing date: 2017.09.25, No.10-2017-0123715; Issued date: 2019.01.30, No.10-1945776-0000

โ€œRecommendation System for Corresponding Messageโ€
Sung-Ju Lee, Joon-Gyum Kim, Taesik Gong, Jun Hyuk Chang, and JeongGil Ko
Korea Patent; Filing date: 2017.08.18, No.10-2017-0104408; Issued date: 2019.05.20, No. 10-1982081-0000


MEDIA

News Articles


Research Videos

MyDJ

ACM CHI '22

MetaSense

ACM SenSys '19

Knocker

ACM IMWUT '19

Media Coverage

ARIRANG TV

Knocker | 2019.11.14

MBC

Knocker | 2019.10.01
Responsive image

KBS

Knocker | 2019.10.01

YTN

Knocker | 2019.10.01

TJB

Knocker | 2019.10.01