metadata

Paper: 2501.13766
license: gpl-3.0
dataset_info: null
configs:
  - config_name: Abstract_algebra
    data_files:
      - split: test
        path: data/Abstract_algebra.json
  - config_name: Arithmetic
    data_files:
      - split: test
        path: data/Arithmetic.json
  - config_name: Calculus_-_single_variable
    data_files:
      - split: test
        path: data/Calculus_-_single_variable.json
  - config_name: Complex_analysis
    data_files:
      - split: test
        path: data/Complex_analysis.json
  - config_name: Financial_mathematics
    data_files:
      - split: test
        path: data/Financial_mathematics.json
  - config_name: Linear_algebra
    data_files:
      - split: test
        path: data/Linear_algebra.json
  - config_name: Probability
    data_files:
      - split: test
        path: data/Probability.json
  - config_name: Statistics
    data_files:
      - split: test
        path: data/Statistics.json
  - config_name: Algebra
    data_files:
      - split: test
        path: data/Algebra.json
  - config_name: Calculus_-_multivariable
    data_files:
      - split: test
        path: data/Calculus_-_multivariable.json
  - config_name: Combinatorics
    data_files:
      - split: test
        path: data/Combinatorics.json
  - config_name: Differential_equations
    data_files:
      - split: test
        path: data/Differential_equations.json
  - config_name: Geometry
    data_files:
      - split: test
        path: data/Geometry.json
  - config_name: Number_theory
    data_files:
      - split: test
        path: data/Number_theory.json
  - config_name: Set_theory_and_logic
    data_files:
      - split: test
        path: data/Set_theory_and_logic.json
  - config_name: Trigonometry
    data_files:
      - split: test
        path: data/Trigonometry.json
task_categories:
  - question-answering
language:
  - en
pretty_name: UGMathBench
size_categories:
  - 10K<n<100K
tags:
  - text
  - math
  - LLMs

UGMathBench: A Diverse and Dynamic Benchmark for Undergraduate-Level Mathematical Reasoning with Large Language Models

UGMathBench is a diverse and dynamic benchmark specifically designed for evaluating undergraduate-level mathematical reasoning with LLMs. UGMathBench comprises 5,062 problems across 16 subjects and 111 topics, featuring 10 distinct answer types. Each problem includes three randomized versions.

An Example to load the data

from datasets import load_dataset
dataset=load_dataset("UGMathBench/ugmathbench", "Trigonometry", split="test")

print(dataset[0])

More details on loading and using the data are on our GitHub page.

If you do find our code helpful or use our benchmark dataset, please cite our paper.

@article{xu2025ugmathbench,
  title={UGMathBench: A Diverse and Dynamic Benchmark for Undergraduate-Level Mathematical Reasoning with Large Language Models},
  author={Xu, Xin and Zhang, Jiaxin and Chen, Tianhao and Chao, Zitong and Hu, Jishan and Yang, Can},
  journal={arXiv preprint arXiv:2501.13766},
  year={2025}
}