Primacy Effect of ChatGPT

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Yiwei, Cai, Yujun, Chen, Muhao, Liang, Yuxuan, Hooi, Bryan
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916244848902144
author Wang, Yiwei
Cai, Yujun
Chen, Muhao
Liang, Yuxuan
Hooi, Bryan
author_facet Wang, Yiwei
Cai, Yujun
Chen, Muhao
Liang, Yuxuan
Hooi, Bryan
contents Instruction-tuned large language models (LLMs), such as ChatGPT, have led to promising zero-shot performance in discriminative natural language understanding (NLU) tasks. This involves querying the LLM using a prompt containing the question, and the candidate labels to choose from. The question-answering capabilities of ChatGPT arise from its pre-training on large amounts of human-written text, as well as its subsequent fine-tuning on human preferences, which motivates us to ask: Does ChatGPT also inherits humans' cognitive biases? In this paper, we study the primacy effect of ChatGPT: the tendency of selecting the labels at earlier positions as the answer. We have two main findings: i) ChatGPT's decision is sensitive to the order of labels in the prompt; ii) ChatGPT has a clearly higher chance to select the labels at earlier positions as the answer. We hope that our experiments and analyses provide additional insights into building more reliable ChatGPT-based solutions. We release the source code at https://github.com/wangywUST/PrimacyEffectGPT.
format Preprint
id arxiv_https___arxiv_org_abs_2310_13206
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Primacy Effect of ChatGPT
Wang, Yiwei
Cai, Yujun
Chen, Muhao
Liang, Yuxuan
Hooi, Bryan
Computation and Language
Artificial Intelligence
Instruction-tuned large language models (LLMs), such as ChatGPT, have led to promising zero-shot performance in discriminative natural language understanding (NLU) tasks. This involves querying the LLM using a prompt containing the question, and the candidate labels to choose from. The question-answering capabilities of ChatGPT arise from its pre-training on large amounts of human-written text, as well as its subsequent fine-tuning on human preferences, which motivates us to ask: Does ChatGPT also inherits humans' cognitive biases? In this paper, we study the primacy effect of ChatGPT: the tendency of selecting the labels at earlier positions as the answer. We have two main findings: i) ChatGPT's decision is sensitive to the order of labels in the prompt; ii) ChatGPT has a clearly higher chance to select the labels at earlier positions as the answer. We hope that our experiments and analyses provide additional insights into building more reliable ChatGPT-based solutions. We release the source code at https://github.com/wangywUST/PrimacyEffectGPT.
title Primacy Effect of ChatGPT
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2310.13206