@inproceedings{agrawal-etal-2025-uncertainty,
title = "Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-{LLM} Systems",
author = "Agrawal, Aakriti and
Aralikatti, Rohith and
Satheesh, Anirudh and
Chakraborty, Souradip and
Bedi, Amrit Singh and
Huang, Furong",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.1367/",
pages = "25090--25098",
ISBN = "979-8-89176-335-7",
abstract = "Large Language Models (LLMs) have demonstrated exceptional capabilities, yet selecting the most reliable response from multiple LLMs remains a challenge, particularly in resource-constrained settings. Existing approaches often depend on costly external verifiers, human evaluators, or self-consistency techniques that require multiple samples from a single model. While multi-LLM systems produce more diverse responses than single models and thus have greater potential, they often underperform compared to single LLM self-consistency. In this work, we propose a calibrated log-likelihood-based selection framework to improve multi-LLM performance. Our approach leverages uncertainty estimation to identify the most confident response while minimizing inference costs. We show that our method outperforms majority voting and exceeds self-consistency performance when using a large number of model calls. Through extensive experiments, we demonstrate improvements of approx. 4{\%}, 3{\%}, and 5{\%} on GSM8K, MMLU, and ARC, respectively, when applying uncertainty-aware selection to multi-LLM systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="agrawal-etal-2025-uncertainty">
<titleInfo>
<title>Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-LLM Systems</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aakriti</namePart>
<namePart type="family">Agrawal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rohith</namePart>
<namePart type="family">Aralikatti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anirudh</namePart>
<namePart type="family">Satheesh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Souradip</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amrit</namePart>
<namePart type="given">Singh</namePart>
<namePart type="family">Bedi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Furong</namePart>
<namePart type="family">Huang</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>Findings of the Association for Computational Linguistics: EMNLP 2025</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">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) have demonstrated exceptional capabilities, yet selecting the most reliable response from multiple LLMs remains a challenge, particularly in resource-constrained settings. Existing approaches often depend on costly external verifiers, human evaluators, or self-consistency techniques that require multiple samples from a single model. While multi-LLM systems produce more diverse responses than single models and thus have greater potential, they often underperform compared to single LLM self-consistency. In this work, we propose a calibrated log-likelihood-based selection framework to improve multi-LLM performance. Our approach leverages uncertainty estimation to identify the most confident response while minimizing inference costs. We show that our method outperforms majority voting and exceeds self-consistency performance when using a large number of model calls. Through extensive experiments, we demonstrate improvements of approx. 4%, 3%, and 5% on GSM8K, MMLU, and ARC, respectively, when applying uncertainty-aware selection to multi-LLM systems.</abstract>
<identifier type="citekey">agrawal-etal-2025-uncertainty</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.1367/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>25090</start>
<end>25098</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-LLM Systems
%A Agrawal, Aakriti
%A Aralikatti, Rohith
%A Satheesh, Anirudh
%A Chakraborty, Souradip
%A Bedi, Amrit Singh
%A Huang, Furong
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F agrawal-etal-2025-uncertainty
%X Large Language Models (LLMs) have demonstrated exceptional capabilities, yet selecting the most reliable response from multiple LLMs remains a challenge, particularly in resource-constrained settings. Existing approaches often depend on costly external verifiers, human evaluators, or self-consistency techniques that require multiple samples from a single model. While multi-LLM systems produce more diverse responses than single models and thus have greater potential, they often underperform compared to single LLM self-consistency. In this work, we propose a calibrated log-likelihood-based selection framework to improve multi-LLM performance. Our approach leverages uncertainty estimation to identify the most confident response while minimizing inference costs. We show that our method outperforms majority voting and exceeds self-consistency performance when using a large number of model calls. Through extensive experiments, we demonstrate improvements of approx. 4%, 3%, and 5% on GSM8K, MMLU, and ARC, respectively, when applying uncertainty-aware selection to multi-LLM systems.
%U https://aclanthology.org/2025.findings-emnlp.1367/
%P 25090-25098
Markdown (Informal)
[Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-LLM Systems](https://aclanthology.org/2025.findings-emnlp.1367/) (Agrawal et al., Findings 2025)
ACL