FarsEval-PKBETS: A new diverse benchmark for evaluating Persian large language models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shamsfard, Mehrnoush, Saaberi, Zahra, manesh, Mostafa Karimi, Hashemi, Seyed Mohammad Hossein, Vatankhah, Zahra, Ramezani, Motahareh, Pourazin, Niki, Zare, Tara, Azimi, Maryam, Chitsaz, Sarina, Khoraminejad, Sama, Mortazavi, Morteza Mahdavi, Chizari, Mohammad Mahdi, Maleki, Sahar, Majd, Seyed Soroush, Masumi, Mostafa, Khoeini, Sayed Ali Musavi, Mohseni, Amir, Alipour, Sogol
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913801114222592
author Shamsfard, Mehrnoush
Saaberi, Zahra
manesh, Mostafa Karimi
Hashemi, Seyed Mohammad Hossein
Vatankhah, Zahra
Ramezani, Motahareh
Pourazin, Niki
Zare, Tara
Azimi, Maryam
Chitsaz, Sarina
Khoraminejad, Sama
Mortazavi, Morteza Mahdavi
Chizari, Mohammad Mahdi
Maleki, Sahar
Majd, Seyed Soroush
Masumi, Mostafa
Khoeini, Sayed Ali Musavi
Mohseni, Amir
Alipour, Sogol
author_facet Shamsfard, Mehrnoush
Saaberi, Zahra
manesh, Mostafa Karimi
Hashemi, Seyed Mohammad Hossein
Vatankhah, Zahra
Ramezani, Motahareh
Pourazin, Niki
Zare, Tara
Azimi, Maryam
Chitsaz, Sarina
Khoraminejad, Sama
Mortazavi, Morteza Mahdavi
Chizari, Mohammad Mahdi
Maleki, Sahar
Majd, Seyed Soroush
Masumi, Mostafa
Khoeini, Sayed Ali Musavi
Mohseni, Amir
Alipour, Sogol
contents Research on evaluating and analyzing large language models (LLMs) has been extensive for resource-rich languages such as English, yet their performance in languages such as Persian has received considerably less attention. This paper introduces FarsEval-PKBETS benchmark, a subset of FarsEval project for evaluating large language models in Persian. This benchmark consists of 4000 questions and answers in various formats, including multiple choice, short answer and descriptive responses. It covers a wide range of domains and tasks,including medicine, law, religion, Persian language, encyclopedic knowledge, human preferences, social knowledge, ethics and bias, text generation, and respecting others' rights. This bechmark incorporates linguistics, cultural, and local considerations relevant to the Persian language and Iran. To ensure the questions are challenging for current LLMs, three models -- Llama3-70B, PersianMind, and Dorna -- were evaluated using this benchmark. Their average accuracy was below 50%, meaning they provided fully correct answers to fewer than half of the questions. These results indicate that current language models are still far from being able to solve this benchmark
format Preprint
id arxiv_https___arxiv_org_abs_2504_14690
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FarsEval-PKBETS: A new diverse benchmark for evaluating Persian large language models
Shamsfard, Mehrnoush
Saaberi, Zahra
manesh, Mostafa Karimi
Hashemi, Seyed Mohammad Hossein
Vatankhah, Zahra
Ramezani, Motahareh
Pourazin, Niki
Zare, Tara
Azimi, Maryam
Chitsaz, Sarina
Khoraminejad, Sama
Mortazavi, Morteza Mahdavi
Chizari, Mohammad Mahdi
Maleki, Sahar
Majd, Seyed Soroush
Masumi, Mostafa
Khoeini, Sayed Ali Musavi
Mohseni, Amir
Alipour, Sogol
Computation and Language
Artificial Intelligence
68T50
I.2.7; E.0
Research on evaluating and analyzing large language models (LLMs) has been extensive for resource-rich languages such as English, yet their performance in languages such as Persian has received considerably less attention. This paper introduces FarsEval-PKBETS benchmark, a subset of FarsEval project for evaluating large language models in Persian. This benchmark consists of 4000 questions and answers in various formats, including multiple choice, short answer and descriptive responses. It covers a wide range of domains and tasks,including medicine, law, religion, Persian language, encyclopedic knowledge, human preferences, social knowledge, ethics and bias, text generation, and respecting others' rights. This bechmark incorporates linguistics, cultural, and local considerations relevant to the Persian language and Iran. To ensure the questions are challenging for current LLMs, three models -- Llama3-70B, PersianMind, and Dorna -- were evaluated using this benchmark. Their average accuracy was below 50%, meaning they provided fully correct answers to fewer than half of the questions. These results indicate that current language models are still far from being able to solve this benchmark
title FarsEval-PKBETS: A new diverse benchmark for evaluating Persian large language models
topic Computation and Language
Artificial Intelligence
68T50
I.2.7; E.0
url https://arxiv.org/abs/2504.14690