@inproceedings{zhao-etal-2026-sage,
title = "{SAGE}: How Well Do {LLM}s Align with Classical {C}hinese Philosophy?",
author = "Zhao, Yuying and
Chen, Sirui and
Liu, Yixin and
Jia, Xuemei and
Min, Lingxi and
Wang, Zheng",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1429/",
pages = "32636--32662",
ISBN = "979-8-89176-459-0",
abstract = "The growing deployment of large language models (LLMs) in socially consequential settings makes it increasingly important to understand the value systems they encode and the moral judgments they produce. However, existing evaluations are largely grounded in Western ethical traditions, leaving non-Western moral perspectives underexplored. To address this gap, we introduce Sage, a benchmark for evaluating LLMs through the lens of Classical Chinese philosophy. Sage is grounded in major Pre-Qin philosophical debates and represents normative alignment in a continuous six-dimensional value space. It includes three task types and is constructed from over 4,000 passages spanning ten philosophical schools. Experiments on eight leading LLMs reveal that models occupy highly similar regions in the proposed value space, exhibit a clear gap between self-declared philosophical identities and benchmark-measured behavior, and show systematically different reasoning patterns across task granularities. Code and data will be released after publication."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2026-sage">
<titleInfo>
<title>SAGE: How Well Do LLMs Align with Classical Chinese Philosophy?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuying</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sirui</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yixin</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuemei</namePart>
<namePart type="family">Jia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lingxi</namePart>
<namePart type="family">Min</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zheng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>The growing deployment of large language models (LLMs) in socially consequential settings makes it increasingly important to understand the value systems they encode and the moral judgments they produce. However, existing evaluations are largely grounded in Western ethical traditions, leaving non-Western moral perspectives underexplored. To address this gap, we introduce Sage, a benchmark for evaluating LLMs through the lens of Classical Chinese philosophy. Sage is grounded in major Pre-Qin philosophical debates and represents normative alignment in a continuous six-dimensional value space. It includes three task types and is constructed from over 4,000 passages spanning ten philosophical schools. Experiments on eight leading LLMs reveal that models occupy highly similar regions in the proposed value space, exhibit a clear gap between self-declared philosophical identities and benchmark-measured behavior, and show systematically different reasoning patterns across task granularities. Code and data will be released after publication.</abstract>
<identifier type="citekey">zhao-etal-2026-sage</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1429/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>32636</start>
<end>32662</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SAGE: How Well Do LLMs Align with Classical Chinese Philosophy?
%A Zhao, Yuying
%A Chen, Sirui
%A Liu, Yixin
%A Jia, Xuemei
%A Min, Lingxi
%A Wang, Zheng
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhao-etal-2026-sage
%X The growing deployment of large language models (LLMs) in socially consequential settings makes it increasingly important to understand the value systems they encode and the moral judgments they produce. However, existing evaluations are largely grounded in Western ethical traditions, leaving non-Western moral perspectives underexplored. To address this gap, we introduce Sage, a benchmark for evaluating LLMs through the lens of Classical Chinese philosophy. Sage is grounded in major Pre-Qin philosophical debates and represents normative alignment in a continuous six-dimensional value space. It includes three task types and is constructed from over 4,000 passages spanning ten philosophical schools. Experiments on eight leading LLMs reveal that models occupy highly similar regions in the proposed value space, exhibit a clear gap between self-declared philosophical identities and benchmark-measured behavior, and show systematically different reasoning patterns across task granularities. Code and data will be released after publication.
%U https://aclanthology.org/2026.emnlp-main.1429/
%P 32636-32662
Markdown (Informal)
[SAGE: How Well Do LLMs Align with Classical Chinese Philosophy?](https://aclanthology.org/2026.emnlp-main.1429/) (Zhao et al., EMNLP 2026)
ACL
- Yuying Zhao, Sirui Chen, Yixin Liu, Xuemei Jia, Lingxi Min, and Zheng Wang. 2026. SAGE: How Well Do LLMs Align with Classical Chinese Philosophy?. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 32636–32662, Budapest, Hungary. Association for Computational Linguistics.