Integrating Large Pre-trained Models into Multimodal Named Entity Recognition with Evidential Fusion

TL;DR

This paper introduces an uncertainty-aware multimodal NER framework that leverages large pre-trained models and evidential fusion to improve prediction trustworthiness and accuracy on social media data.

Contribution

It integrates uncertainty estimation with large pre-trained models in MNER, providing a novel evidential fusion mechanism for better interpretability and performance.

Findings

Outperforms baseline methods on two datasets

Achieves state-of-the-art accuracy in MNER

Enhances prediction trustworthiness through uncertainty modeling

Abstract

Multimodal Named Entity Recognition (MNER) is a crucial task for information extraction from social media platforms such as Twitter. Most current methods rely on attention weights to extract information from both text and images but are often unreliable and lack interpretability. To address this problem, we propose incorporating uncertainty estimation into the MNER task, producing trustworthy predictions. Our proposed algorithm models the distribution of each modality as a Normal-inverse Gamma distribution, and fuses them into a unified distribution with an evidential fusion mechanism, enabling hierarchical characterization of uncertainties and promotion of prediction accuracy and trustworthiness. Additionally, we explore the potential of pre-trained large foundation models in MNER and propose an efficient fusion approach that leverages their robust feature representations. Experiments on two datasets demonstrate that our proposed method outperforms the baselines and achieves new state-of-the-art performance.