Back to search

University of Waterloo

Uncovering the Reliability and Consistency of AI Language Models: A Systematic Study

Abstract

dc:description.abstract

Large Language Models (LLMs) have rapidly advanced, becoming general-purpose assistants and creative partners. Despite their widespread use, LLMs exhibit significant vulnerabilities to prompt variations and struggle with task understanding, leading to inconsistencies and factual inaccuracies in their responses. Traditional Natural Language Processing (NLP) benchmarks often overlook nuances in LLM behavior and reliability. This thesis addresses this gap by curating a dataset across six categories: Fact, Conspiracy, Controversy, Misconception, Stereotype, and Fiction. We rigorously define LLMs' factual accuracy, consistency, and robustness to prompt variations using diverse response formats and question variations, and evaluate these on 37 models. Our findings reveal LLMs' volatility and unreliability, particularly in the Controversy and Misconception categories, where conflicting training data impedes performance. Additionally, we explore LLMs' ability to generate coherent fictional narratives, probing their ability to retain and effectively utilize factual information, a critical requirement for creative tasks like story generation. While LLMs offer versatile applications, their reliability hinges on addressing challenges in prompt understanding and response consistency, emphasizing the need for ongoing research to enhance their performance across diverse tasks and applications.

Degree

thesis:*
Grantor dc:publisher
University of Waterloo
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Khatun, Aisha
Advisor dc:contributor.advisor
  • Brown, Dan

Subjects

dc:subject × 16

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/10012/20847

Chain of custody

source
Harvested from
University of Waterloo
Base URL
dspacemainprd01.lib.uwaterloo.ca/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Khatun, Aisha. Uncovering the Reliability and Consistency of AI Language Models: A Systematic Study. University of Waterloo, 2024. https://hdl.handle.net/10012/20847