# Confidence-Credibility Aware Weighted Ensembles of Small LLMs Outperform Large LLMs in Emotion Detection

arXiv: https://arxiv.org/abs/2512.17630 · HTML: https://mlsanity.com/paper/2512.17630/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 38 references that are in the map.
- First posted on arXiv 19 Dec 2025.

## Builds on (most cited first, 15 of 15)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [The Falcon Series of Open Language Models](https://mlsanity.com/paper/2311.16867/) (2023)
- [Crowdsourcing a Word-Emotion Association Lexicon](https://mlsanity.com/paper/1308.6297/) (2013)
- [A Unified Theory of Diversity in Ensemble Learning](https://mlsanity.com/paper/2301.03962/) (2023)
- [DICE: Diversity in Deep Ensembles via Conditional Redundancy Adversarial Estimation](https://mlsanity.com/paper/2101.05544/) (2021)
- [A Systematic Evaluation of LLM Strategies for Mental Health Text Analysis: Fine-tuning vs. Prompt Engineering vs. RAG](https://mlsanity.com/paper/2503.24307/) (2025)
- [Ensembling over Classifiers: a Bias-Variance Perspective](https://mlsanity.com/paper/2206.10566/) (2022)
- [Fast and reliable uncertainty quantification with neural network ensembles for industrial image classification](https://mlsanity.com/paper/2403.10182/) (2024)
- [Emotion Classification In Software Engineering Texts: A Comparative Analysis of Pre-trained Transformers Language Models](https://mlsanity.com/paper/2401.10845/) (2024)

## Built on by (most cited first, 1 of 1)

- [A Self Consistency Based Reranking for Narrative Question Answering](https://mlsanity.com/paper/2606.15741/) (2026)

## Abstract (arXiv)

This paper introduces a confidence-weighted, credibility-aware ensemble framework for text-based emotion detection, inspired by Condorcet's Jury Theorem (CJT). Unlike conventional ensembles that often rely on homogeneous architectures, our approach combines architecturally diverse small transformer-based large language models (sLLMs) - BERT, RoBERTa, DistilBERT, DeBERTa, and ELECTRA, each fully fine-tuned for emotion classification. To preserve error diversity, we minimize parameter convergence while taking advantage of the unique biases of each model. A dual-weighted voting mechanism integrates both global credibility (validation F1 score) and local confidence (instance-level probability) to dynamically weight model contributions. Experiments on the DAIR-AI dataset demonstrate that our credibility-confidence ensemble achieves a macro F1 score of 93.5 percent, surpassing state-of-the-art benchmarks and significantly outperforming large-scale LLMs, including Falcon, Mistral, Qwen, and Phi, even after task-specific Low-Rank Adaptation (LoRA). With only 595M parameters in total, our small LLMs ensemble proves more parameter-efficient and robust than models up to 7B parameters, establishing that carefully designed ensembles of small, fine-tuned models can outperform much larger LLMs in specialized natural language processing (NLP) tasks such as emotion detection.