Large Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds (Minsky, 1988), we propose ReConcile, a multi-model multi-agent framework designed as a round table conference among diverse LLM agents. ReConcile enhances collaborative reasoning between LLM agents via multiple rounds of discussion, learning to convince other agents to improve their answers, and employing a confidence-weighted voting mechanism that leads to a better consensus. In each round, ReConcile initiates discussion between agents via a ‘discussion prompt’ that consists of (a) grouped answers and explanations generated by each agent in the previous round, (b) their confidence scores, and (c) demonstrations of answer-rectifying human explanations, used for convincing other agents. Experiments on seven benchmarks demonstrate that ReConcile significantly improves LLMs’ reasoning – both individually and as a team – surpassing prior single-agent and multi-agent baselines by up to 11.4% and even outperforming GPT-4 on three datasets. ReConcile also flexibly incorporates different combinations of agents, including API-based, open-source, and domain-specific models, leading to an 8% improvement on MATH. Finally, we analyze the individual components of ReConcile, demonstrating that the diversity originating from different models is critical to its superior performance.
@inproceedings{chen-etal-2024-reconcile, title = "{R}e{C}oncile: Round-Table Conference Improves Reasoning via Consensus among Diverse {LLM}s", author = "Chen, Justin and Saha, Swarnadeep and Bansal, Mohit", editor = "Ku, Lun-Wei and Martins, Andre and Srikumar, Vivek", booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = aug, year = "2024", address = "Bangkok, Thailand", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2024.acl-long.381/", doi = "10.18653/v1/2024.acl-long.381", pages = "7066--7085", abstract = "Large Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds (Minsky, 1988), we propose ReConcile, a multi-model multi-agent framework designed as a round table conference among diverse LLM agents. ReConcile enhances collaborative reasoning between LLM agents via multiple rounds of discussion, learning to convince other agents to improve their answers, and employing a confidence-weighted voting mechanism that leads to a better consensus. In each round, ReConcile initiates discussion between agents via a {\textquoteleft}discussion prompt' that consists of (a) grouped answers and explanations generated by each agent in the previous round, (b) their confidence scores, and (c) demonstrations of answer-rectifying human explanations, used for convincing other agents. Experiments on seven benchmarks demonstrate that ReConcile significantly improves LLMs' reasoning {--} both individually and as a team {--} surpassing prior single-agent and multi-agent baselines by up to 11.4{\%} and even outperforming GPT-4 on three datasets. ReConcile also flexibly incorporates different combinations of agents, including API-based, open-source, and domain-specific models, leading to an 8{\%} improvement on MATH. Finally, we analyze the individual components of ReConcile, demonstrating that the diversity originating from different models is critical to its superior performance."}
<?xml version="1.0" encoding="UTF-8"?><modsCollection xmlns="http://www.loc.gov/mods/v3"><mods ID="chen-etal-2024-reconcile"> <titleInfo> <title>ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs</title> </titleInfo> <name type="personal"> <namePart type="given">Justin</namePart> <namePart type="family">Chen</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Swarnadeep</namePart> <namePart type="family">Saha</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Mohit</namePart> <namePart type="family">Bansal</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <originInfo> <dateIssued>2024-08</dateIssued> </originInfo> <typeOfResource>text</typeOfResource> <relatedItem type="host"> <titleInfo> <title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title> </titleInfo> <name type="personal"> <namePart type="given">Lun-Wei</namePart> <namePart type="family">Ku</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Andre</namePart> <namePart type="family">Martins</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Vivek</namePart> <namePart type="family">Srikumar</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <originInfo> <publisher>Association for Computational Linguistics</publisher> <place> <placeTerm type="text">Bangkok, Thailand</placeTerm> </place> </originInfo> <genre authority="marcgt">conference publication</genre> </relatedItem> <abstract>Large Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds (Minsky, 1988), we propose ReConcile, a multi-model multi-agent framework designed as a round table conference among diverse LLM agents. ReConcile enhances collaborative reasoning between LLM agents via multiple rounds of discussion, learning to convince other agents to improve their answers, and employing a confidence-weighted voting mechanism that leads to a better consensus. In each round, ReConcile initiates discussion between agents via a ‘discussion prompt’ that consists of (a) grouped answers and explanations generated by each agent in the previous round, (b) their confidence scores, and (c) demonstrations of answer-rectifying human explanations, used for convincing other agents. Experiments on seven benchmarks demonstrate that ReConcile significantly improves LLMs’ reasoning – both individually and as a team – surpassing prior single-agent and multi-agent baselines by up to 11.4% and even outperforming GPT-4 on three datasets. ReConcile also flexibly incorporates different combinations of agents, including API-based, open-source, and domain-specific models, leading to an 8% improvement on MATH. Finally, we analyze the individual components of ReConcile, demonstrating that the diversity originating from different models is critical to its superior performance.</abstract> <identifier type="citekey">chen-etal-2024-reconcile</identifier> <identifier type="doi">10.18653/v1/2024.acl-long.381</identifier> <location> <url>https://aclanthology.org/2024.acl-long.381/</url> </location> <part> <date>2024-08</date> <extent unit="page"> <start>7066</start> <end>7085</end> </extent> </part></mods></modsCollection>
%0 Conference Proceedings%T ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs%A Chen, Justin%A Saha, Swarnadeep%A Bansal, Mohit%Y Ku, Lun-Wei%Y Martins, Andre%Y Srikumar, Vivek%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)%D 2024%8 August%I Association for Computational Linguistics%C Bangkok, Thailand%F chen-etal-2024-reconcile%X Large Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds (Minsky, 1988), we propose ReConcile, a multi-model multi-agent framework designed as a round table conference among diverse LLM agents. ReConcile enhances collaborative reasoning between LLM agents via multiple rounds of discussion, learning to convince other agents to improve their answers, and employing a confidence-weighted voting mechanism that leads to a better consensus. In each round, ReConcile initiates discussion between agents via a ‘discussion prompt’ that consists of (a) grouped answers and explanations generated by each agent in the previous round, (b) their confidence scores, and (c) demonstrations of answer-rectifying human explanations, used for convincing other agents. Experiments on seven benchmarks demonstrate that ReConcile significantly improves LLMs’ reasoning – both individually and as a team – surpassing prior single-agent and multi-agent baselines by up to 11.4% and even outperforming GPT-4 on three datasets. ReConcile also flexibly incorporates different combinations of agents, including API-based, open-source, and domain-specific models, leading to an 8% improvement on MATH. Finally, we analyze the individual components of ReConcile, demonstrating that the diversity originating from different models is critical to its superior performance.%R 10.18653/v1/2024.acl-long.381%U https://aclanthology.org/2024.acl-long.381/%U https://doi.org/10.18653/v1/2024.acl-long.381%P 7066-7085
[ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs](https://aclanthology.org/2024.acl-long.381/) (Chen et al., ACL 2024)