Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Loya, Manikanta, Sinha, Divya Anand, Futrell, Richard
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2312.17476
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916076427673600
author Loya, Manikanta
Sinha, Divya Anand
Futrell, Richard
author_facet Loya, Manikanta
Sinha, Divya Anand
Futrell, Richard
contents The advancement of Large Language Models (LLMs) has led to their widespread use across a broad spectrum of tasks including decision making. Prior studies have compared the decision making abilities of LLMs with those of humans from a psychological perspective. However, these studies have not always properly accounted for the sensitivity of LLMs' behavior to hyperparameters and variations in the prompt. In this study, we examine LLMs' performance on the Horizon decision making task studied by Binz and Schulz (2023) analyzing how LLMs respond to variations in prompts and hyperparameters. By experimenting on three OpenAI language models possessing different capabilities, we observe that the decision making abilities fluctuate based on the input prompts and temperature settings. Contrary to previous findings language models display a human-like exploration exploitation tradeoff after simple adjustments to the prompt.
format Preprint
id arxiv_https___arxiv_org_abs_2312_17476
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Exploring the Sensitivity of LLMs' Decision-Making Capabilities: Insights from Prompt Variation and Hyperparameters
Loya, Manikanta
Sinha, Divya Anand
Futrell, Richard
Computation and Language
The advancement of Large Language Models (LLMs) has led to their widespread use across a broad spectrum of tasks including decision making. Prior studies have compared the decision making abilities of LLMs with those of humans from a psychological perspective. However, these studies have not always properly accounted for the sensitivity of LLMs' behavior to hyperparameters and variations in the prompt. In this study, we examine LLMs' performance on the Horizon decision making task studied by Binz and Schulz (2023) analyzing how LLMs respond to variations in prompts and hyperparameters. By experimenting on three OpenAI language models possessing different capabilities, we observe that the decision making abilities fluctuate based on the input prompts and temperature settings. Contrary to previous findings language models display a human-like exploration exploitation tradeoff after simple adjustments to the prompt.
title Exploring the Sensitivity of LLMs' Decision-Making Capabilities: Insights from Prompt Variation and Hyperparameters
topic Computation and Language
url https://arxiv.org/abs/2312.17476