Towards Human-AI Collaborative Information Synthesis

PhD Thesis Proposal Defence


Title: "Towards Human-AI Collaborative Information Synthesis"

by

Mr. Chengbo ZHENG


Abstract:

Synthesizing information into a holistic understanding or a coherent narrative 
is a common necessity. For example, researchers need to synthesize related work 
to ground their studies. The information synthesis process typically involves 
gathering information from various sources, distilling insights, and organizing 
them coherently. Although foundation models, such as large language models 
(LLMs), offer significant potential for automating this process, they still 
exhibit deficiencies like bias and instability. Additionally, users often seek 
control over the synthesis process to integrate their unique contexts or to 
learn from the process.

This thesis investigates the human-AI collaboration approach to information 
synthesis, aiming to optimally leverage the capabilities of foundation models 
while maintaining user agency. Specifically, in the first part of the thesis, 
we adopt a user-centered design approach to explore how human-AI collaboration 
can be integrated into common information synthesis tasks. First, we designed 
NB2Slides, an AI-assisted presentation slide creation system that supports data 
scientists in synthesizing their technical work for communication with 
stakeholders. Our second work focuses on supporting knowledge integration for 
interdisciplinary research by using LLMs to guide the orientation of 
information gathering and rapid screening of relevant literature. The second 
part of this thesis examines how to enable human-AI collaboration in 
synthesizing newly emerged information, such as students' use of AI tools in 
course projects. Lacking an empirical understanding of the needs and challenges 
in synthesizing such information, we initially conducted co-design future 
workshops to explore students' motivations and attitudes. Based on the insights 
gathered, we will design an intelligent tool to help students synthesize their 
AI usage and project traces into learning diaries, thus supporting their 
reflection on their AI-enhanced learning journeys. In summary, this thesis aims 
to contribute design knowledge on forming human-AI workflows for information 
synthesis and an empirical understanding of human-AI interaction to inform 
future interactive AI system design.


Date:                   Tuesday, 11 June 2024

Time:                   2:00pm - 4:00pm

Venue:                  Room 3494
                        Lifts 25/26

Committee Members:      Dr. Xiaojuan Ma (Supervisor)
                        Prof. Pedro Sander (Chairperson)
                        Dr. Yangqiu Song
                        Dr. Mingming Fan