AROMA: Adaptive Orchestration for Robust and Cost-Efficient Multi-Agent LLM Systems

AROMA: Adaptive Orchestration for Robust and Cost-Efficient Multi-Agent LLM Systems | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 20 March 2026 V1 Latest version Share on AROMA: Adaptive Orchestration for Robust and Cost-Efficient Multi-Agent LLM Systems Authors : Tianyu Yin 0009-0009-2880-6624 [email protected] and Jingyi Jia Authors Info & Affiliations https://doi.org/10.22541/au.177403077.77403552/v1 117 views 78 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Multi-Agent Large Language Model Systems (MAS) confront significant challenges stemming from systemic design flaws, inter-agent misalignments, and prohibitive operational costs. These systems frequently offer only modest performance gains, or even exhibit setbacks, while incurring substantial increases in token consumption due to prevalent failure modes like improper task decomposition and information overload. To address these critical issues, we propose Adaptive Orchestration for Robust Multi-Agent LLM Systems (AROMA), a novel framework designed to dynamically perceive, diagnose, and adaptively orchestrate multi-agent collaboration. AROMA incorporates capabilities for real-time failure identification, intelligent adjustment of system parameters, roles, and communication strategies, and optimization for efficient task completion with minimal overhead. Through extensive experiments on complex benchmarks, AROMA demonstrates enhanced task success rates and a substantial reduction in average token cost compared to existing baselines, alongside a significant mitigation of collaboration failure modes. Our findings confirm AROMA's efficacy in improving robustness, efficiency, and generalizability, paving the way for more reliable and economically sustainable multi-agent LLM deployments. Supplementary Material File (aroma.pdf) Download 2.35 MB Information & Authors Information Version history V1 Version 1 20 March 2026 Copyright This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License Keywords adaptive orchestration collaboration efficiency multi-agent llm systems robustness Authors Affiliations Tianyu Yin 0009-0009-2880-6624 [email protected] Zhongnan University of Economics and Law View all articles by this author Jingyi Jia Zhongnan University of Economics and Law View all articles by this author Metrics & Citations Metrics Article Usage 117 views 78 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Tianyu Yin, Jingyi Jia. AROMA: Adaptive Orchestration for Robust and Cost-Efficient Multi-Agent LLM Systems. Authorea . 20 March 2026. DOI: https://doi.org/10.22541/au.177403077.77403552/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.177403077.77403552/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe22f6f8ec70700',t:'MTc3OTE4NDA5OQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();