Bayesian Network Fusion of Large Language Models for Sentiment Analysis
This work addresses the problem of improving interpretability and robustness in sentiment analysis for financial domains, though it is incremental as it builds on existing LLMs with a fusion method.
The paper tackles the challenges of transparency, cost, and inconsistency in domain-specific large language models (LLMs) for sentiment analysis by proposing the Bayesian network LLM fusion (BNLF) framework, which integrates predictions from three LLMs through a probabilistic mechanism, resulting in consistent gains of about six percent in accuracy over baseline LLMs across three financial corpora.
Large language models (LLMs) continue to advance, with an increasing number of domain-specific variants tailored for specialised tasks. However, these models often lack transparency and explainability, can be costly to fine-tune, require substantial prompt engineering, yield inconsistent results across domains, and impose significant adverse environmental impact due to their high computational demands. To address these challenges, we propose the Bayesian network LLM fusion (BNLF) framework, which integrates predictions from three LLMs, including FinBERT, RoBERTa, and BERTweet, through a probabilistic mechanism for sentiment analysis. BNLF performs late fusion by modelling the sentiment predictions from multiple LLMs as probabilistic nodes within a Bayesian network. Evaluated across three human-annotated financial corpora with distinct linguistic and contextual characteristics, BNLF demonstrates consistent gains of about six percent in accuracy over the baseline LLMs, underscoring its robustness to dataset variability and the effectiveness of probabilistic fusion for interpretable sentiment classification.