Saved in:
Bibliographic Details
Main Authors: Meister, Nicole, Guestrin, Carlos, Hashimoto, Tatsunori
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.05403
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915009760591872
author Meister, Nicole
Guestrin, Carlos
Hashimoto, Tatsunori
author_facet Meister, Nicole
Guestrin, Carlos
Hashimoto, Tatsunori
contents Language models (LMs) are increasingly used as simulacra for people, yet their ability to match the distribution of views of a specific demographic group and be \textit{distributionally aligned} remains uncertain. This notion of distributional alignment is complex, as there is significant variation in the types of attributes that are simulated. Prior works have underexplored the role of three critical variables -- the question domain, steering method, and distribution expression method -- which motivates our contribution of a benchmark explicitly addressing these dimensions. We construct a dataset expanding beyond political values, create human baselines for this task, and evaluate the extent to which an LM can align with a particular group's opinion distribution to inform design choices of such simulation systems. Our analysis reveals open problems regarding if, and how, LMs can be used to simulate humans, and that LLMs can more accurately describe the opinion distribution than simulate such distributions.
format Preprint
id arxiv_https___arxiv_org_abs_2411_05403
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Benchmarking Distributional Alignment of Large Language Models
Meister, Nicole
Guestrin, Carlos
Hashimoto, Tatsunori
Computation and Language
Artificial Intelligence
Language models (LMs) are increasingly used as simulacra for people, yet their ability to match the distribution of views of a specific demographic group and be \textit{distributionally aligned} remains uncertain. This notion of distributional alignment is complex, as there is significant variation in the types of attributes that are simulated. Prior works have underexplored the role of three critical variables -- the question domain, steering method, and distribution expression method -- which motivates our contribution of a benchmark explicitly addressing these dimensions. We construct a dataset expanding beyond political values, create human baselines for this task, and evaluate the extent to which an LM can align with a particular group's opinion distribution to inform design choices of such simulation systems. Our analysis reveals open problems regarding if, and how, LMs can be used to simulate humans, and that LLMs can more accurately describe the opinion distribution than simulate such distributions.
title Benchmarking Distributional Alignment of Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2411.05403