AgriEval: A Comprehensive Chinese Agricultural Benchmark for Large Language Models

Lian Yan, Haotian Wang, Chen Tang, Haifeng Liu, Tianyang Sun, Liangliang Liu, Yi Guan, Jingchi Jiang · 2025

Read AgriEval: A Comprehensive Chinese Agricultural Benchmark for Large Language Models by Lian Yan, Haotian Wang, Chen Tang, Haifeng Liu, Tianyang Sun, Liangliang Liu, Yi Guan, Jingchi Jiang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-21773.txt.

Description

Lian Yan and a team of co-authors introduce AgriEval in a 2025 arXiv paper. AgriEval is a test set for judging how well large language models handle farm topics in Chinese. Such models lack good tests in this field, which slows their use in farming.

The set covers six broad farm areas and 29 sub-areas, and it tests recall, grasp, reasoning, and writing. Its questions come from university exams and coursework. It holds many multiple-choice questions and many open-ended ones, the largest set of its kind so far. The team tested 51 models, both open and commercial. Most scored under 60 percent, which shows much room to grow. They also study what affects scores and suggest ways to improve.

Catalog record

CREATORLian Yan, Haotian Wang, Chen Tang, Haifeng Liu, Tianyang Sun, Liangliang Liu, Yi Guan, Jingchi Jiang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE102 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-21773

Browse the full library · About the Tower of Babel