Single and Multi Truth Data Fusion using Large Language Models

2026-06-26

AI Summary

This study addresses the challenge of identifying conflicting attribute values in data fusion under both single-truth and multi-truth scenarios. It presents the first systematic exploration of large language models (LLMs) for this task, proposing a unified prompt engineering framework that seamlessly handles both truth paradigms in an end-to-end manner. The framework encompasses zero-shot, one-shot, and domain-related or domain-agnostic strategies. Experimental results on three benchmark datasets demonstrate that the proposed approach significantly outperforms traditional unsupervised methods such as DART and LTM, thereby validating the effectiveness and versatility of LLMs in data fusion. The implementation code has been made publicly available.

Abstract

Data fusion, also known as truth discovery, is a data integration problem that aims to determine the correct value or set of values for each attribute of an object when presented with potentially conflicting values from multiple sources. Data fusion tasks belong to two main categories: single-truth scenarios, where each attribute has only one correct value, and multi-truth scenarios, where multiple values can be valid simultaneously. This paper investigates the use of Large Language Models (LLMs) in data fusion tasks for tabular data. Various prompting strategies, encompassing both single-truth and multi-truth scenarios, are investigated empirically. Domain-dependent, domain-independent, zero-shot and one-shot prompts are evaluated on three different benchmark datasets. Experimental results demonstrate that LLM-based approaches outperform traditional unsupervised truth discovery methods, such as DART and LTM, across all datasets. The codebase of this study has been made publicly available on GitHub.

Problem

• data fusion
• truth discovery
• single-truth
• multi-truth
• tabular data

Innovation

• Large Language Models
• Data Fusion
• Truth Discovery
• Prompt Engineering
• Tabular Data