Jaeseong Lee
Development
texas, United States
Skills
Machine Learning (ML)
About
JAESAEONG LEE's skills align with IT R&D Professionals (Information and Communication Technology). JAESAEONG also has skills associated with Programmers (Information and Communication Technology). JAESAEONG LEE has 7 years of work experience.
View more
Work Experience
Graduate Research Assistant
Software Engineering Group
August 2021 - Present
- • Develping an automated mutation-based framework to assess the arithmetic reasoning capability of Large Language Models. - Producing in-context examples by implementing context-aware mutation technique to mitigate reliance of external dataset and in-context example retrieval model. - Optimizing quality of in-context examples by selecting subset of in-context examples that maximize the inter- and intraconsistency of the LLM responses over modalities. • Developed a toolchain, ALiCT, the first automated tool that generates test cases according to linguistic capabilities in multiple NLP applications (e.g. sentiment analysis and hate speech detection). • Estimating the impact of unsuccessful test cases to determine probabilistic error associations among the test cases.
Teaching Assistant (TA), University
Texas at Austin
August 2021 - December 2021
- Introduction of Engineering Design (EE364D) Aug 2020 - Dec 2020
EE397K
August 2019 - December 2019
Senior Project Design
Health Science
January 2019 - May 2019
- Data Analysis for Health Science (SDS302) Jan 2018 - May 2018
Teaching Assistant (TA), University
Texas at Dallas
January 2023 - May 2023
- Software Architecture and Design (SE4352) Aug 2022 - Dec 2022
Graduate Research Assistant
Software Engineering Group
August 2018 - May 2021
- • Analyzed naturalness of hardware description languages (HDL) and developed an automated code completion model for the hardware description languages. • Developed machine learning-based software engineering tools including code summarization and code completion. - Developed a generative sequence-to-sequence model from the large HDL dataset in the domain of code suggestion and code completion tasks.
Human-Robot Interaction
August 2017 - December 2017
- Introduction of Engineering Design (EE465D) Aug 2014 - Dec 2014
Algorithm Development Engineer intern, Advanced Driver Assistant System
FaradayFuture
May 2017 - August 2017
- • Participated and developed multi-sensor (Camera, Lidar, Radar and USS) based object detection algorithms. • Built algorithm module testing framework for autonomous parking system.
Engineering intern
Spirent Communications
June 2016 - August 2016
- • Tested PEVQ VQA algorithm for usage of chromatic software for video experience evaluation. • Implemented VMAF video quality assessment algorithm into the chromatic video experience evaluation software.
Graduate Research Assistant. Supervisor
Laboratory for Image and Video Engineering (LIVE Lab)
November 2015 - December 2016
- • Developed a learning-based video scene encoding and decoding scheme using GAN-based adversarial autoencoder. - Developed a video compression model, compressing video frame into down-sampled edge-frame and reconstructing it into full frame using conditional GAN-based decoder. - Adversarial training decoder with a discriminator, which judges whether its inputs are authentic or synthetic images by competing with one another. • Analyzed video BLIINDS, a video quality assessment algorithm, for developing a video archiving system. INDUSTRIAL EXPERIENCE
Education
University of Texas at Dallas
Ph.D. in Computer Science.
August 2021 - Present
University of Texas
M.S. in Electrical Engineering.
August 2014 - May 2017
