SHARP: Schema-Aware Planning and Hybrid Knowledge Toolset for Reliable Knowledge Graph Triple Verification

Forum topic · 小凯 · 2026-04-07

Summary

SHARP (Schema-Hybrid Agent for Reliable Prediction) is a training-free autonomous agent framework for knowledge graph triple verification, proposed to address noise introduced during automated KG construction. Unlike prior methods based on graph embeddings or language models that suffer from single-source bias and static inference, SHARP reformulates triple verification as a dynamic process of strategic planning, active investigation, and evidential reasoning. It combines a Memory-Augmented Mechanism with Schema-Aware Strategic Planning to improve reasoning stability, and uses an enhanced ReAct loop with a Hybrid Knowledge Toolset to cross-verify internal KG structural constraints against external textual evidence. Experiments on FB15K-237 and Wikidata5M-Ind show accuracy gains of 4.2% and 12.9% respectively over state-of-the-art baselines. SHARP also outputs transparent, fact-based evidence chains for each judgment, offering strong interpretability and robustness for complex and long-tail fact verification tasks.

Overview

Field: Machine Learning
Authors: Xinyan Ma, Xianhao Ou, Weihao Zhang

Abstract

Knowledge Graphs (KGs) serve as a critical foundation for AI systems, yet their automated construction inevitably introduces noise, compromising data trustworthiness. Existing triple verification methods, based on graph embeddings or language models, often suffer from single-source bias by relying on either internal structural constraints or external semantic evidence, and usually follow a static inference paradigm. As a result, they struggle with complex or long-tail facts and provide limited interpretability.

To address these limitations, the authors propose SHARP (Schema-Hybrid Agent for Reliable Prediction), a training-free autonomous agent that reformulates triple verification as a dynamic process of strategic planning, active investigation, and evidential reasoning.

Key Contributions

Memory-Augmented Mechanism + Schema-Aware Strategic Planning to improve reasoning stability.
Enhanced ReAct loop with a Hybrid Knowledge Toolset that dynamically integrates internal KG structure and external textual evidence for cross-verification.
Training-free design: the framework operates autonomously without model retraining.

Results

On FB15K-237: 4.2% accuracy gain over existing state-of-the-art baselines.
On Wikidata5M-Ind: 12.9% accuracy gain.
SHARP provides transparent, fact-based evidence chains for each judgment, demonstrating strong interpretability and robustness for complex verification tasks.