SLIDE: Sliding Localized Information for Document Extraction
This addresses the challenge of constructing accurate knowledge graphs from long documents and low-resource languages, which is incremental as it builds on existing retrieval methods.
The paper tackled the problem of degraded performance in knowledge graph extraction from long texts and low-resource languages by introducing SLIDE, a chunking method that uses overlapping windows to retain contextual information, resulting in improvements such as a 24% increase in entity extraction and 39% in relationship extraction for English, and up to 82% for Afrikaans.
Constructing accurate knowledge graphs from long texts and low-resource languages is challenging, as large language models (LLMs) experience degraded performance with longer input chunks. This problem is amplified in low-resource settings where data scarcity hinders accurate entity and relationship extraction. Contextual retrieval methods, while improving retrieval accuracy, struggle with long documents. They truncate critical information in texts exceeding maximum context lengths of LLMs, significantly limiting knowledge graph construction. We introduce SLIDE (Sliding Localized Information for Document Extraction), a chunking method that processes long documents by generating local context through overlapping windows. SLIDE ensures that essential contextual information is retained, enhancing knowledge graph extraction from documents exceeding LLM context limits. It significantly improves GraphRAG performance, achieving a 24% increase in entity extraction and a 39% improvement in relationship extraction for English. For Afrikaans, a low-resource language, SLIDE achieves a 49% increase in entity extraction and an 82% improvement in relationship extraction. Furthermore, it improves upon state-of-the-art in question-answering metrics such as comprehensiveness, diversity and empowerment, demonstrating its effectiveness in multilingual and resource-constrained settings.