View Proposal

Proposer
Mike Chantler
Title
How to increase trust in AI summarisation
Goal
To provide interactive visualisations that help users quickly understand the reasoning and evidence behind AI summarisation of documents.
Description
AI sometimes lie and make stuff up - so users are often highly suspicious of AI summaries of research papers and other documents. This project would provide graphical means of examining the evidence and sources used for AI summarisation so that users could quickly understand the validity of, confidence in, and evidence used for AI summarisation of scientific papers.
Resources
Background
Url
Difficulty Level
Moderate
Ethical Approval
Full
Number Of Students
2
Supervisor
Mike Chantler
Keywords
visualisation, llms, nlp
Degrees
Bachelor of Science in Computer Systems
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
BSc Information Systems with Data Analytics
Back to List