@inproceedings{wang-etal-2026-meeseeks-multilingual,
title = "Meeseeks-Multilingual: A Multi-Turn, Multi-Constraint Benchmark evaluating Multilingual Instruction-Following and Self-Correction",
author = "Wang, Jiaming and
Chen, Hefei and
Tang, Zhe and
Jin, Zehao and
Zhuo, Yanxin and
Liao, Ruicheng and
Ma, Jinrong and
Li, Xiaoyu and
Cao, Xuezhi",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.2349/",
pages = "50505--50518",
ISBN = "979-8-89176-462-0",
abstract = "Large Language Models (LLMs) fail to reliably follow multi-constraint instructions across languages, and the failure rate rises when regional linguistic rules interact with general constraints. To address this challenge, we present Meeseeks-Multilingual, a multi-turn, feedback-driven benchmark that extends Meeseeks to multilingual and cross-regional evaluation. The benchmark separates general constraints (e.g., word counts, keyword rules) from regional constraints (e.g., Japanese kana requirements) so that the two failure modes can be measured independently. Meeseeks-Multilingual automatically identifies unmet constraints and provides targeted feedback to test iterative self-correction. All data are produced by trained annotators with no LLM involvement in data generation. Experiments across mainstream commercial and open-source LLMs show that even with iterative feedback, models consistently fail to meet all multilingual constraints. Aggregate model rankings stabilize by Round 5, which lets the benchmark be run at reduced cost for overall comparisons, although regional rankings settle more slowly and need more rounds. Meeseeks-Multilingual has been open-sourced on https://github.com/meituan-longcat/Meeseeks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-meeseeks-multilingual">
<titleInfo>
<title>Meeseeks-Multilingual: A Multi-Turn, Multi-Constraint Benchmark evaluating Multilingual Instruction-Following and Self-Correction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiaming</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hefei</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhe</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zehao</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanxin</namePart>
<namePart type="family">Zhuo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruicheng</namePart>
<namePart type="family">Liao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinrong</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoyu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuezhi</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) fail to reliably follow multi-constraint instructions across languages, and the failure rate rises when regional linguistic rules interact with general constraints. To address this challenge, we present Meeseeks-Multilingual, a multi-turn, feedback-driven benchmark that extends Meeseeks to multilingual and cross-regional evaluation. The benchmark separates general constraints (e.g., word counts, keyword rules) from regional constraints (e.g., Japanese kana requirements) so that the two failure modes can be measured independently. Meeseeks-Multilingual automatically identifies unmet constraints and provides targeted feedback to test iterative self-correction. All data are produced by trained annotators with no LLM involvement in data generation. Experiments across mainstream commercial and open-source LLMs show that even with iterative feedback, models consistently fail to meet all multilingual constraints. Aggregate model rankings stabilize by Round 5, which lets the benchmark be run at reduced cost for overall comparisons, although regional rankings settle more slowly and need more rounds. Meeseeks-Multilingual has been open-sourced on https://github.com/meituan-longcat/Meeseeks.</abstract>
<identifier type="citekey">wang-etal-2026-meeseeks-multilingual</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.2349/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>50505</start>
<end>50518</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Meeseeks-Multilingual: A Multi-Turn, Multi-Constraint Benchmark evaluating Multilingual Instruction-Following and Self-Correction
%A Wang, Jiaming
%A Chen, Hefei
%A Tang, Zhe
%A Jin, Zehao
%A Zhuo, Yanxin
%A Liao, Ruicheng
%A Ma, Jinrong
%A Li, Xiaoyu
%A Cao, Xuezhi
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F wang-etal-2026-meeseeks-multilingual
%X Large Language Models (LLMs) fail to reliably follow multi-constraint instructions across languages, and the failure rate rises when regional linguistic rules interact with general constraints. To address this challenge, we present Meeseeks-Multilingual, a multi-turn, feedback-driven benchmark that extends Meeseeks to multilingual and cross-regional evaluation. The benchmark separates general constraints (e.g., word counts, keyword rules) from regional constraints (e.g., Japanese kana requirements) so that the two failure modes can be measured independently. Meeseeks-Multilingual automatically identifies unmet constraints and provides targeted feedback to test iterative self-correction. All data are produced by trained annotators with no LLM involvement in data generation. Experiments across mainstream commercial and open-source LLMs show that even with iterative feedback, models consistently fail to meet all multilingual constraints. Aggregate model rankings stabilize by Round 5, which lets the benchmark be run at reduced cost for overall comparisons, although regional rankings settle more slowly and need more rounds. Meeseeks-Multilingual has been open-sourced on https://github.com/meituan-longcat/Meeseeks.
%U https://aclanthology.org/2026.findings-emnlp.2349/
%P 50505-50518
Markdown (Informal)
[Meeseeks-Multilingual: A Multi-Turn, Multi-Constraint Benchmark evaluating Multilingual Instruction-Following and Self-Correction](https://aclanthology.org/2026.findings-emnlp.2349/) (Wang et al., Findings 2026)
ACL
- Jiaming Wang, Hefei Chen, Zhe Tang, Zehao Jin, Yanxin Zhuo, Ruicheng Liao, Jinrong Ma, Xiaoyu Li, and Xuezhi Cao. 2026. Meeseeks-Multilingual: A Multi-Turn, Multi-Constraint Benchmark evaluating Multilingual Instruction-Following and Self-Correction. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 50505–50518, Budapest, Hungary. Association for Computational Linguistics.