Towards Efficient and Domain-aware Adaptation of Large Foundation Models

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Towards Efficient and Domain-aware Adaptation of Large Foundation
Models"

By

Mr. Shizhe DIAO


Abstract:

In the burgeoning realm of artificial intelligence, foundation models stand out
as a pivotal advancement. Demonstrating an unmatched ability in human-like text
comprehension and generation, these models have set new standards in a wide
array of natural language processing applications. Their general-purpose nature
allows impressive performance across diverse tasks, yet domain-aware adaptation
can further amplify their efficacy. The challenge is to fine-tune these models
to specific domains efficiently and effectively. In this thesis, we offer an
in-depth exploration into taming language and vision-language models to
particular domains. Our focus is on architectural modifications, training
strategies, and prompting methods that enhance large foundation model
performance in specific domains while ensuring the adaptation is
resourceefficient, scalable, and effective.


Date:                   Monday, 27 November 2023

Time:                   10:00am - 12:00noon

Venue:                  Room 5510
                        Lifts 25/26

Chairman:               Prof. Ross MURCH (ECE)

Committee Members:      Prof. Tong ZHANG (Supervisor)
                        Prof. Raymond WONG
                        Prof. Xiaofang ZHOU
                        Prof. Kani CHEN (MATH)
                        Prof. Irwin KING (CUHK)


**** ALL are Welcome ****