GraphCheck: Multi-Path Fact-Checking with Entity-Relationship Graphs

2025-02-28

AI Summary

To address the challenge of fact-checking complex claims requiring multi-hop reasoning, this paper proposes GraphCheck—a novel framework that models claims as entity-relation graphs and performs end-to-end verification via explicit and implicit multi-path relational reasoning among entities. Methodologically, we introduce a two-stage DP-GraphCheck: Stage I employs direct prompting to filter noisy inputs; Stage II jointly integrates graph neural network representations, multi-hop path reasoning, and evidence retrieval for verification. GraphCheck establishes the first graph-structured, multi-path fact-checking paradigm, significantly enhancing robustness and generalization. On the HOVER and EX-FEVER benchmarks, it achieves a 7.2% absolute accuracy improvement over prior state-of-the-art methods on multi-hop tasks. Moreover, DP-GraphCheck is modular and plug-and-play, enabling seamless integration into existing fact-checking pipelines without architectural modifications.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Automated fact-checking aims to assess the truthfulness of text based on relevant evidence, yet verifying complex claims requiring multi-hop reasoning remains a significant challenge. We propose GraphCheck, a novel framework that converts claims into entity-relationship graphs for comprehensive verification. By identifying relation between explicit entities and latent entities across multiple paths, GraphCheck enhances the adaptability and robustness of verification. Furthermore, we introduce DP-GraphCheck, a two-stage variant that improves performance by incorporating direct prompting as an initial filtering step. Experiments on the HOVER and EX-FEVER datasets show that our approach outperforms existing methods, particularly in multi-hop reasoning tasks. Furthermore, our two-stage framework generalizes well to other fact-checking pipelines, demonstrating its versatility.

Problem

• Automated fact-checking of complex claims requiring multi-hop reasoning.
• Enhancing verification adaptability and robustness using entity-relationship graphs.
• Improving performance with a two-stage framework incorporating direct prompting.

Innovation

• Converts claims into entity-relationship graphs
• Identifies relations across multiple paths
• Introduces DP-GraphCheck with direct prompting