T-GRAG: A Dynamic GraphRAG Framework for Resolving Temporal Conflicts and Redundancy in Knowledge Retrieval

2025-08-03

AI Summary

Existing GraphRAG methods neglect the temporal dynamics of knowledge, resulting in temporal ambiguity, time-insensitive retrieval, and semantic redundancy. To address this, we propose T-GRAG—a novel framework that introduces time-aware dynamic knowledge graph construction into GraphRAG for the first time. T-GRAG enables cross-temporal multi-hop reasoning via temporal query decomposition, three-layer interactive retrieval, and source-text denoising extraction. It innovatively integrates timestamped graph generation with dynamic graph neural networks to support time-sensitive question answering over long documents. We further release Time-LongQA, the first benchmark for temporal question answering over corporate annual reports. Experiments demonstrate that T-GRAG significantly outperforms state-of-the-art RAG and GraphRAG baselines in both retrieval accuracy and response relevance, validating the critical importance of explicitly modeling knowledge evolution for temporal QA.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Large language models (LLMs) have demonstrated strong performance in natural language generation but remain limited in knowle- dge-intensive tasks due to outdated or incomplete internal knowledge. Retrieval-Augmented Generation (RAG) addresses this by incorporating external retrieval, with GraphRAG further enhancing performance through structured knowledge graphs and multi-hop reasoning. However, existing GraphRAG methods largely ignore the temporal dynamics of knowledge, leading to issues such as temporal ambiguity, time-insensitive retrieval, and semantic redundancy. To overcome these limitations, we propose Temporal GraphRAG (T-GRAG), a dynamic, temporally-aware RAG framework that models the evolution of knowledge over time. T-GRAG consists of five key components: (1) a Temporal Knowledge Graph Generator that creates time-stamped, evolving graph structures; (2) a Temporal Query Decomposition mechanism that breaks complex temporal queries into manageable sub-queries; (3) a Three-layer Interactive Retriever that progressively filters and refines retrieval across temporal subgraphs; (4) a Source Text Extractor to mitigate noise; and (5) a LLM-based Generator that synthesizes contextually and temporally accurate responses. We also introduce Time-LongQA, a novel benchmark dataset based on real-world corporate annual reports, designed to test temporal reasoning across evolving knowledge. Extensive experiments show that T-GRAG significantly outperforms prior RAG and GraphRAG baselines in both retrieval accuracy and response relevance under temporal constraints, highlighting the necessity of modeling knowledge evolution for robust long-text question answering. Our code is publicly available on the T-GRAG

Problem

• Addresses temporal ambiguity in knowledge retrieval
• Resolves time-insensitive retrieval in GraphRAG methods
• Reduces semantic redundancy in dynamic knowledge graphs

Innovation

• Temporal Knowledge Graph Generator for evolving knowledge
• Three-layer Interactive Retriever for temporal subgraphs
• Time-LongQA benchmark for temporal reasoning