MoVa: Towards Generalizable Classification of Human Morals and Values

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Ziyu, Sun, Junfei, Li, Chenxi, Nguyen, Tuan Dung, Yao, Jing, Yi, Xiaoyuan, Xie, Xing, Tan, Chenhao, Xie, Lexing
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911182577729536
author Chen, Ziyu
Sun, Junfei
Li, Chenxi
Nguyen, Tuan Dung
Yao, Jing
Yi, Xiaoyuan
Xie, Xing
Tan, Chenhao
Xie, Lexing
author_facet Chen, Ziyu
Sun, Junfei
Li, Chenxi
Nguyen, Tuan Dung
Yao, Jing
Yi, Xiaoyuan
Xie, Xing
Tan, Chenhao
Xie, Lexing
contents Identifying human morals and values embedded in language is essential to empirical studies of communication. However, researchers often face substantial difficulty navigating the diversity of theoretical frameworks and data available for their analysis. Here, we contribute MoVa, a well-documented suite of resources for generalizable classification of human morals and values, consisting of (1) 16 labeled datasets and benchmarking results from four theoretically-grounded frameworks; (2) a lightweight LLM prompting strategy that outperforms fine-tuned models across multiple domains and frameworks; and (3) a new application that helps evaluate psychological surveys. In practice, we specifically recommend a classification strategy, all@once, that scores all related concepts simultaneously, resembling the well-known multi-label classifier chain. The data and methods in MoVa can facilitate many fine-grained interpretations of human and machine communication, with potential implications for the alignment of machine behavior.
format Preprint
id arxiv_https___arxiv_org_abs_2509_24216
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MoVa: Towards Generalizable Classification of Human Morals and Values
Chen, Ziyu
Sun, Junfei
Li, Chenxi
Nguyen, Tuan Dung
Yao, Jing
Yi, Xiaoyuan
Xie, Xing
Tan, Chenhao
Xie, Lexing
Computation and Language
Computers and Society
Identifying human morals and values embedded in language is essential to empirical studies of communication. However, researchers often face substantial difficulty navigating the diversity of theoretical frameworks and data available for their analysis. Here, we contribute MoVa, a well-documented suite of resources for generalizable classification of human morals and values, consisting of (1) 16 labeled datasets and benchmarking results from four theoretically-grounded frameworks; (2) a lightweight LLM prompting strategy that outperforms fine-tuned models across multiple domains and frameworks; and (3) a new application that helps evaluate psychological surveys. In practice, we specifically recommend a classification strategy, all@once, that scores all related concepts simultaneously, resembling the well-known multi-label classifier chain. The data and methods in MoVa can facilitate many fine-grained interpretations of human and machine communication, with potential implications for the alignment of machine behavior.
title MoVa: Towards Generalizable Classification of Human Morals and Values
topic Computation and Language
Computers and Society
url https://arxiv.org/abs/2509.24216