@inproceedings{son-etal-2025-arena,
title = "Arena-lite: Efficient and Reliable Large Language Model Evaluation via Tournament-Based Direct Comparisons",
author = "Son, Seonil and
Oh, Ju-Min and
Jin, Heegon and
Jang, Cheolhun and
Jeong, Jeongbeom and
Kim, KunTae",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.360/",
pages = "7068--7086",
ISBN = "979-8-89176-332-6",
abstract = "As Large Language Models (LLMs) expand across domains, LLM judges have become essential for systems evaluation. Current benchmarks typically compare system outputs against baselines.This baseline-mediated approach, though convenient, yields lower reliability than direct comparison between systems.We propose Arena-Lite which integrates tournament structure on top of head-to-head comparison.The application of a tournament structure and direct comparison eliminates the need for baseline outputs, reduces the number of required comparisons, and allows higher reliability in system rankings.We conducted two experiments: (1) controlled stochastic modeling and (2) empirical validation with a real LLM judge. Those experiments collectively demonstrate that Arena-Lite consistently achieves higher reliability with fewer comparisons, even with smaller datasets or weaker judges.We release an easy-to-use web demonstration and code to foster adoption of Arena-Lite, streamlining model selection across research and industry communities. Arena-Lite demo and code are available on https://huggingface.co/spaces/NCSOFT/ArenaLite"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="son-etal-2025-arena">
<titleInfo>
<title>Arena-lite: Efficient and Reliable Large Language Model Evaluation via Tournament-Based Direct Comparisons</title>
</titleInfo>
<name type="personal">
<namePart type="given">Seonil</namePart>
<namePart type="family">Son</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ju-Min</namePart>
<namePart type="family">Oh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heegon</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cheolhun</namePart>
<namePart type="family">Jang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jeongbeom</namePart>
<namePart type="family">Jeong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">KunTae</namePart>
<namePart type="family">Kim</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>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>As Large Language Models (LLMs) expand across domains, LLM judges have become essential for systems evaluation. Current benchmarks typically compare system outputs against baselines.This baseline-mediated approach, though convenient, yields lower reliability than direct comparison between systems.We propose Arena-Lite which integrates tournament structure on top of head-to-head comparison.The application of a tournament structure and direct comparison eliminates the need for baseline outputs, reduces the number of required comparisons, and allows higher reliability in system rankings.We conducted two experiments: (1) controlled stochastic modeling and (2) empirical validation with a real LLM judge. Those experiments collectively demonstrate that Arena-Lite consistently achieves higher reliability with fewer comparisons, even with smaller datasets or weaker judges.We release an easy-to-use web demonstration and code to foster adoption of Arena-Lite, streamlining model selection across research and industry communities. Arena-Lite demo and code are available on https://huggingface.co/spaces/NCSOFT/ArenaLite</abstract>
<identifier type="citekey">son-etal-2025-arena</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.360/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>7068</start>
<end>7086</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Arena-lite: Efficient and Reliable Large Language Model Evaluation via Tournament-Based Direct Comparisons
%A Son, Seonil
%A Oh, Ju-Min
%A Jin, Heegon
%A Jang, Cheolhun
%A Jeong, Jeongbeom
%A Kim, KunTae
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F son-etal-2025-arena
%X As Large Language Models (LLMs) expand across domains, LLM judges have become essential for systems evaluation. Current benchmarks typically compare system outputs against baselines.This baseline-mediated approach, though convenient, yields lower reliability than direct comparison between systems.We propose Arena-Lite which integrates tournament structure on top of head-to-head comparison.The application of a tournament structure and direct comparison eliminates the need for baseline outputs, reduces the number of required comparisons, and allows higher reliability in system rankings.We conducted two experiments: (1) controlled stochastic modeling and (2) empirical validation with a real LLM judge. Those experiments collectively demonstrate that Arena-Lite consistently achieves higher reliability with fewer comparisons, even with smaller datasets or weaker judges.We release an easy-to-use web demonstration and code to foster adoption of Arena-Lite, streamlining model selection across research and industry communities. Arena-Lite demo and code are available on https://huggingface.co/spaces/NCSOFT/ArenaLite
%U https://aclanthology.org/2025.emnlp-main.360/
%P 7068-7086
Markdown (Informal)
[Arena-lite: Efficient and Reliable Large Language Model Evaluation via Tournament-Based Direct Comparisons](https://aclanthology.org/2025.emnlp-main.360/) (Son et al., EMNLP 2025)
ACL