Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp01fb494c76q
Title: Chatting with GPT: Pragmatic-Pedagogic Dialogue Games
Authors: Omeike, Etiosa
Advisors: Fernandez Fisac, Jaime
Department: Electrical and Computer Engineering
Certificate Program: Robotics & Intelligent Systems Program
Class Year: 2024
Abstract: The human-robot interaction community has shown growing interest in Large Language Model (LLM) powered agents like ChatGPT and LLaMA for their advanced communication skills. However, these fine-tuned LLMs often lack strategic conversation abilities. They tend to provide lengthy, verbose responses without considering their usefulness to humans, and they may not effectively gather information from the humans they assist. Generally, LLMs don't match human conversational skills, a shortfall possibly due to their training. Typically, LLMs are fine-tuned through a one-shot process based on human or AI-evaluated prompt rankings, which might not support sustained, helpful human interaction. This thesis investigates the use of game-theoretic tools for more effective conversational LLM fine-tuning, with a specific focus on the issues of developing computationally tractable algorithms with appealing theoretical properties (eg, convergence) to enable LLMs to handle value alignment even in the face of uncertainty.
URI: http://arks.princeton.edu/ark:/88435/dsp01fb494c76q
Type of Material: Princeton University Senior Theses
Language: en
Appears in Collections:Electrical and Computer Engineering, 1932-2024
Robotics and Intelligent Systems Program

Files in This Item:
File Description SizeFormat 
OMEIKE-ETIOSA-THESIS.pdf316.57 kBAdobe PDF    Request a copy


Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.