ChatGPT and its robustness, fairness, trustworthiness and impact

The swift progress of AI has brought about a new age of LLMs, with models such as ChatGPT-4 leading the way in these advancements. With the integration of different types of inputs like text, images, and other data, managing robustness, fairness, trustworthiness and hallucinations in models becomes...

Full description

Saved in:
Bibliographic Details
Main Author: Muhammad Akmal Bin Rahmat
Other Authors: Alex Chichung Kot
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/181710
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-181710
record_format dspace
spelling sg-ntu-dr.10356-1817102024-12-20T15:45:37Z ChatGPT and its robustness, fairness, trustworthiness and impact Muhammad Akmal Bin Rahmat Alex Chichung Kot School of Electrical and Electronic Engineering EACKOT@ntu.edu.sg Engineering ChatGPT The swift progress of AI has brought about a new age of LLMs, with models such as ChatGPT-4 leading the way in these advancements. With the integration of different types of inputs like text, images, and other data, managing robustness, fairness, trustworthiness and hallucinations in models becomes more difficult when the generated information is not accurately connected to the input provided. This report delves into examining the resilience, equity, and reliability of ChatGPT-4. The main goal of this study is to analyze the occurrence of inconsistent, unfair and untrustworthy answers and hallucinations in ChatGPT-4 when handling multimodal inputs and to examine the effectiveness of Factually Augmented Reinforcement Learning from Human Feedback (RLHF) as a possible remedy. The research includes an in-depth examination of the pertinent literature on LLMs, RLHF, and multimodal hallucinations, paired with an empirical evaluation utilizing a dataset of images and accompanying textual replies produced by ChatGPT-4. Important discoveries show that although ChatGPT-4 shows prowess in some respects like recognizing its limitations and giving precise quantitative information, it struggles with complex comparative analyses and providing thorough contextual details. The use of Factually Augmented RLHF offers potential in tackling these problems by including more factual information and clearer reward signals, leading to a decrease in hallucinations and improvement in overall model performance. This report enhances our knowledge of AI robustness and fairness by exploring the unique difficulties of multimodal hallucinations and assessing possible ways to address them. The results highlight the importance of continued research and improvement in AI systems to guarantee their dependability and credibility in practical settings. Bachelor's degree 2024-12-16T02:20:45Z 2024-12-16T02:20:45Z 2024 Final Year Project (FYP) Muhammad Akmal Bin Rahmat (2024). ChatGPT and its robustness, fairness, trustworthiness and impact. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/181710 https://hdl.handle.net/10356/181710 en P3014-231 application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering
ChatGPT
spellingShingle Engineering
ChatGPT
Muhammad Akmal Bin Rahmat
ChatGPT and its robustness, fairness, trustworthiness and impact
description The swift progress of AI has brought about a new age of LLMs, with models such as ChatGPT-4 leading the way in these advancements. With the integration of different types of inputs like text, images, and other data, managing robustness, fairness, trustworthiness and hallucinations in models becomes more difficult when the generated information is not accurately connected to the input provided. This report delves into examining the resilience, equity, and reliability of ChatGPT-4. The main goal of this study is to analyze the occurrence of inconsistent, unfair and untrustworthy answers and hallucinations in ChatGPT-4 when handling multimodal inputs and to examine the effectiveness of Factually Augmented Reinforcement Learning from Human Feedback (RLHF) as a possible remedy. The research includes an in-depth examination of the pertinent literature on LLMs, RLHF, and multimodal hallucinations, paired with an empirical evaluation utilizing a dataset of images and accompanying textual replies produced by ChatGPT-4. Important discoveries show that although ChatGPT-4 shows prowess in some respects like recognizing its limitations and giving precise quantitative information, it struggles with complex comparative analyses and providing thorough contextual details. The use of Factually Augmented RLHF offers potential in tackling these problems by including more factual information and clearer reward signals, leading to a decrease in hallucinations and improvement in overall model performance. This report enhances our knowledge of AI robustness and fairness by exploring the unique difficulties of multimodal hallucinations and assessing possible ways to address them. The results highlight the importance of continued research and improvement in AI systems to guarantee their dependability and credibility in practical settings.
author2 Alex Chichung Kot
author_facet Alex Chichung Kot
Muhammad Akmal Bin Rahmat
format Final Year Project
author Muhammad Akmal Bin Rahmat
author_sort Muhammad Akmal Bin Rahmat
title ChatGPT and its robustness, fairness, trustworthiness and impact
title_short ChatGPT and its robustness, fairness, trustworthiness and impact
title_full ChatGPT and its robustness, fairness, trustworthiness and impact
title_fullStr ChatGPT and its robustness, fairness, trustworthiness and impact
title_full_unstemmed ChatGPT and its robustness, fairness, trustworthiness and impact
title_sort chatgpt and its robustness, fairness, trustworthiness and impact
publisher Nanyang Technological University
publishDate 2024
url https://hdl.handle.net/10356/181710
_version_ 1819113016730320896