@inproceedings{liu-etal-2025-highmath,
title = "{H}igh{MATH}: Evaluating Math Reasoning of Large Language Models in Breadth and Depth",
author = "Liu, Yan and
Zhang, Minghui and
Xiong, Bojian and
Xiao, Yifan and
Sun, Yinong and
Mei, Yating and
Zeng, Longyu and
Yang, Jingchao and
Wang, Yang and
Xiong, Deyi",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.542/",
doi = "10.18653/v1/2025.findings-emnlp.542",
pages = "10241--10253",
ISBN = "979-8-89176-335-7",
abstract = "With the rapid development of large language models (LLMs) in math reasoning, the accuracy of models on existing math benchmarks has gradually approached 90{\%} or even higher. More challenging math benchmarks are hence urgently in need to satisfy the increasing evaluation demands. To bridge this gap, we propose HighMATH. Problems in HighMATH are collected according to 3 criteria: problem complexity, knowledge domain diversity and fine-grained annotations. We collect 5,293 problems from Chinese senior high school mathematics exams published in 2024, covering 8 subjects and 7 levels of difficulty, with each problem involving an average of more than 2.4 knowledge points. We conduct a thorough evaluation of latest LLMs on the curated HighMATH, including o1-like models. Evaluation results demonstrate that the accuracy of advanced LLMs on HighMATH is significantly lower than that on previous math reasoning benchmarks. This gap even exceeds 30{\%}. Our results also suggest that properly trained smaller LLMs may have great potential in math reasoning. Our data is available at https://github.com/tjunlp-lab/HighMATH."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2025-highmath">
<titleInfo>
<title>HighMATH: Evaluating Math Reasoning of Large Language Models in Breadth and Depth</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghui</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bojian</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yifan</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yinong</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yating</namePart>
<namePart type="family">Mei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Longyu</namePart>
<namePart type="family">Zeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingchao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deyi</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>With the rapid development of large language models (LLMs) in math reasoning, the accuracy of models on existing math benchmarks has gradually approached 90% or even higher. More challenging math benchmarks are hence urgently in need to satisfy the increasing evaluation demands. To bridge this gap, we propose HighMATH. Problems in HighMATH are collected according to 3 criteria: problem complexity, knowledge domain diversity and fine-grained annotations. We collect 5,293 problems from Chinese senior high school mathematics exams published in 2024, covering 8 subjects and 7 levels of difficulty, with each problem involving an average of more than 2.4 knowledge points. We conduct a thorough evaluation of latest LLMs on the curated HighMATH, including o1-like models. Evaluation results demonstrate that the accuracy of advanced LLMs on HighMATH is significantly lower than that on previous math reasoning benchmarks. This gap even exceeds 30%. Our results also suggest that properly trained smaller LLMs may have great potential in math reasoning. Our data is available at https://github.com/tjunlp-lab/HighMATH.</abstract>
<identifier type="citekey">liu-etal-2025-highmath</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.542</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.542/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>10241</start>
<end>10253</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HighMATH: Evaluating Math Reasoning of Large Language Models in Breadth and Depth
%A Liu, Yan
%A Zhang, Minghui
%A Xiong, Bojian
%A Xiao, Yifan
%A Sun, Yinong
%A Mei, Yating
%A Zeng, Longyu
%A Yang, Jingchao
%A Wang, Yang
%A Xiong, Deyi
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F liu-etal-2025-highmath
%X With the rapid development of large language models (LLMs) in math reasoning, the accuracy of models on existing math benchmarks has gradually approached 90% or even higher. More challenging math benchmarks are hence urgently in need to satisfy the increasing evaluation demands. To bridge this gap, we propose HighMATH. Problems in HighMATH are collected according to 3 criteria: problem complexity, knowledge domain diversity and fine-grained annotations. We collect 5,293 problems from Chinese senior high school mathematics exams published in 2024, covering 8 subjects and 7 levels of difficulty, with each problem involving an average of more than 2.4 knowledge points. We conduct a thorough evaluation of latest LLMs on the curated HighMATH, including o1-like models. Evaluation results demonstrate that the accuracy of advanced LLMs on HighMATH is significantly lower than that on previous math reasoning benchmarks. This gap even exceeds 30%. Our results also suggest that properly trained smaller LLMs may have great potential in math reasoning. Our data is available at https://github.com/tjunlp-lab/HighMATH.
%R 10.18653/v1/2025.findings-emnlp.542
%U https://aclanthology.org/2025.findings-emnlp.542/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.542
%P 10241-10253
Markdown (Informal)
[HighMATH: Evaluating Math Reasoning of Large Language Models in Breadth and Depth](https://aclanthology.org/2025.findings-emnlp.542/) (Liu et al., Findings 2025)
ACL
- Yan Liu, Minghui Zhang, Bojian Xiong, Yifan Xiao, Yinong Sun, Yating Mei, Longyu Zeng, Jingchao Yang, Yang Wang, and Deyi Xiong. 2025. HighMATH: Evaluating Math Reasoning of Large Language Models in Breadth and Depth. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 10241–10253, Suzhou, China. Association for Computational Linguistics.