Skip to main navigation Skip to search Skip to main content

Self-Ensemble: Mitigating Confidence Distortion for Large Language Models

  • Zicheng Xu
  • , Guanchu Wang
  • , Guangyao Zheng
  • , Yu Neng Chuang
  • , Alexander Szalay
  • , Xia Hu
  • , Vladimir Braverman

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Although Large Language Models (LLMs) perform well in general fields, they exhibit a confidence distortion problem on multi-choice question-answering (MCQA), particularly as the number of answer choices increases. Specifically, on MCQA with many choices, LLMs suffer from under-confidence in correct predictions and over-confidence in incorrect ones, leading to a substantially degraded performance. To solve this problem, we propose Self-Ensemble in this work. Our method splits the choices into several groups and ensembles LLM predictions across these groups to reach a final decision. The advantage of Self-Ensemble is its plug-and-play nature, where it can be integrated into existing LLM architecture based on a designed attention mask and positional encoding, without requiring labeled datasets for parameter tuning. Experimental results on three LLMs and datasets demonstrate that Self-Ensemble comprehensively addresses the confidence distortion problem of LLMs, outperforming standard inference as well as baseline methods. The source code is available at https://github. com/ZichengXu/Self-Ensemble.

Original languageEnglish
Title of host publicationEMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025
EditorsChristos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
PublisherAssociation for Computational Linguistics (ACL)
Pages16603-16615
Number of pages13
ISBN (Electronic)9798891763357
DOIs
StatePublished - 1 Jan 2025
Externally publishedYes
Event30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025 - Suzhou, China
Duration: 4 Nov 20259 Nov 2025

Publication series

NameEMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025

Conference

Conference30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025
Country/TerritoryChina
CitySuzhou
Period4/11/259/11/25

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Information Systems
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'Self-Ensemble: Mitigating Confidence Distortion for Large Language Models'. Together they form a unique fingerprint.

Cite this