AGENTICT$^2$S:Robust Text-to-SPARQL via Agentic Collaborative Reasoning over Heterogeneous Knowledge Graphs for the Circular Economy

2025-08-03

AI Summary

Existing text-to-SPARQL approaches rely on large-scale domain-specific fine-tuning or single-knowledge-graph settings, limiting generalization to low-resource scenarios and complex cross-source knowledge graph question answering. To address semantic querying over heterogeneous, decentralized knowledge graphs in circular economy applications (e.g., classification, process, and emission graphs), we propose a modular multi-agent framework that decomposes the task into retrieval, generation, and verification stages. Our method introduces a weak-to-strong alignment strategy and a two-stage symbolic verification mechanism—including counterfactual consistency checking—to jointly ensure structural and semantic consistency across disparate graphs. Experiments on real-world circular economy knowledge graphs demonstrate a 17.3% improvement in execution accuracy, a 25.4% gain in triple-level F₁ score, and a 46.4% reduction in average prompt length. The framework significantly enhances robustness and interpretability in low-resource, multi-graph environments.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Question answering over heterogeneous knowledge graphs (KGQA) involves reasoning across diverse schemas, incomplete alignments, and distributed data sources. Existing text-to-SPARQL approaches rely on large-scale domain-specific fine-tuning or operate within single-graph settings, limiting their generalizability in low-resource domains and their ability to handle queries spanning multiple graphs. These challenges are particularly relevant in domains such as the circular economy, where information about classifications, processes, and emissions is distributed across independently curated knowledge graphs (KGs). We present AgenticT$^2$S, a modular framework that decomposes KGQA into subtasks managed by specialized agents responsible for retrieval, query generation, and verification. A scheduler assigns subgoals to different graphs using weak-to-strong alignment strategies. A two-stage verifier detects structurally invalid and semantically underspecified queries through symbolic validation and counterfactual consistency checks. Experiments on real-world circular economy KGs demonstrate that AgenticT$^2$S improves execution accuracy by 17.3% and triple level F$_1$ by 25.4% over the best baseline, while reducing the average prompt length by 46.4%. These results demonstrate the benefits of agent-based schema-aware reasoning for scalable KGQA and support decision-making in sustainability domains through robust cross-graph reasoning.

Problem

• Handles queries across multiple heterogeneous knowledge graphs
• Improves generalizability in low-resource circular economy domains
• Detects invalid and underspecified queries via verification

Innovation

• Modular framework with specialized agents
• Weak-to-strong alignment for multi-graph queries
• Two-stage verifier for query validation