mlsanityData to 14 Oct 2026

A Simple Ensemble Strategy for LLM Inference: Towards More Stable Text Classification

arXiv 2504.18884 ↗First posted 26 Apr 2025cs.CL cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

With the advance of large language models (LLMs), LLMs have been utilized for the various tasks. However, the issues of variability and reproducibility of results from each trial of LLMs have been largely overlooked in existing literature while actual human annotation uses majority voting to resolve disagreements among annotators. Therefore, this study introduces the straightforward ensemble strategy to a sentiment analysis using LLMs. As the results, we demonstrate that the ensemble of multiple inference using medium-sized LLMs produces more robust and accurate results than using a large model with a single attempt with reducing RMSE by 18.6%.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 18 of its 48 references that are in the map. First posted on arXiv 26 Apr 2025.

Builds on

18 of 48 references.

References · 18

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  4. 2017Deep reinforcement learning from human preferencescited by 3,912
  5. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  6. 2016Bag of Tricks for Efficient Text Classificationcited by 876
  7. 2023ChatGPT Outperforms Crowd-Workers for Text-Annotation Taskscited by 870
  8. 2022Is GPT-3 a Good Data Annotator?cited by 208
  9. 2018Deep Learning for Sentiment Analysis : A Surveycited by 139
  10. 2023CodeGen2: Lessons for Training LLMs on Programming and Natural Languagescited by 125
  11. 2016Yelp Dataset Challenge: Review Rating Predictioncited by 84
  12. 2023Automated Annotation with Generative AI Requires Validationcited by 76
6 more
  1. 2024Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaborationcited by 67
  2. 2024Designing Heterogeneous LLM Agents for Financial Sentiment Analysiscited by 48
  3. 2018An Improved Text Sentiment Classification Model Using TF-IDF and Next Word Negationcited by 16
  4. 2020Yelp Review Rating Prediction: Machine Learning and Deep Learning Modelscited by 5
  5. 2024Dynamic Sentiment Analysis with Local Large Language Models using Majority Voting: A Study on Factors Affecting Restaurant Evaluationcited by 5
  6. 2024Multimodal Deep Learning of Word-of-Mouth Text and Demographics to Predict Customer Rating: Handling Consumer Heterogeneity in Marketingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 5.

8 papers.

  1. 2025Harnessing Consistency for Robust Test-Time LLM Ensemblecited by 10
  2. 2026ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inferencecited by 5
  3. 2026Multi-Agent Taint Specification Extraction for Vulnerability Detectioncited by 3
  4. 2026RAL2M: Retrieval Augmented Learning-To-Match Against Hallucination in Compliance-Guaranteed Service Systemscited by 2
  5. 2026Automated Root-Cause Subclassification and No-Code Fix Generation for Invalid Bug Reportscited by 1
  6. 2025Reference Points in LLM Sentiment Analysis: The Role of Structured Contextcited by 0
  7. 2025Stable LLM Ensemble: Interaction between Example Representativeness and Diversitycited by 0
  8. 2026Can Urban Blight Be Accessed with Vision-language Models: A Case Study in Detroitcited by 0