ClaimPKG: Enhancing Claim Verification via Pseudo-Subgraph Generation with Lightweight Specialized LLM

TL;DR

ClaimPKG is a novel framework that enhances claim verification by generating pseudo-subgraphs with a lightweight LLM to effectively leverage knowledge graphs, leading to improved accuracy and zero-shot generalization.

Contribution

It introduces a lightweight LLM for pseudo-subgraph generation and an end-to-end integration with KGs, advancing claim verification methods.

Findings

Achieves 9%-12% higher accuracy than baselines on FactKG.

Demonstrates effective zero-shot transfer to unstructured datasets.

Outperforms existing methods across multiple LLM backbones.

Abstract

Integrating knowledge graphs (KGs) to enhance the reasoning capabilities of large language models (LLMs) is an emerging research challenge in claim verification. While KGs provide structured, semantically rich representations well-suited for reasoning, most existing verification methods rely on unstructured text corpora, limiting their ability to effectively leverage KGs. Additionally, despite possessing strong reasoning abilities, modern LLMs struggle with multi-step modular pipelines and reasoning over KGs without adaptation. To address these challenges, we propose ClaimPKG, an end-to-end framework that seamlessly integrates LLM reasoning with structured knowledge from KGs. Specifically, the main idea of ClaimPKG is to employ a lightweight, specialized LLM to represent the input claim as pseudo-subgraphs, guiding a dedicated subgraph retrieval module to identify relevant KG subgraphs. These retrieved subgraphs are then processed by a general-purpose LLM to produce the final verdict and justification. Extensive experiments on the FactKG dataset demonstrate that ClaimPKG achieves state-of-the-art performance, outperforming strong baselines in this research field by 9%-12% accuracy points across multiple categories. Furthermore, ClaimPKG exhibits zero-shot generalizability to unstructured datasets such as HoVer and FEVEROUS, effectively combining structured knowledge from KGs with LLM reasoning across various LLM backbones.