Beyond Chunks and Graphs: Retrieval-Augmented Generation through Triplet-Driven Thinking

ICLR 2026 Conference Submission20184 Authors

Keywords: retrieval-augmented generation, chain-of-thought, knowledge graphs

TL;DR: We propose T$^2$RAG, a novel method that combines the advantages of both approaches while significantly reducing cost. T$^2$RAG leverages the inherent thinking capabilities of LLMs to expand questions into traceable triples containing question marks.

Abstract: Retrieval-augmented generation (RAG) is critical for reducing hallucinations and incorporating external knowledge into Large Language Models (LLMs). However, advanced RAG systems face a trade-off between performance and efficiency. Multi-round RAG approaches achieve strong reasoning but incur excessive LLM calls and token costs, while Graph RAG methods suffer from computationally expensive, error-prone graph construction and retrieval redundancy. To address these challenges, we propose T$^2$RAG, a novel framework that operates on a simple, graph-free knowledge base of atomic triplets. T$^2$RAG leverages an LLM to decompose questions into searchable triplets with placeholders, which it then iteratively resolves by retrieving evidence from the triplet database. Empirical results show that T$^2$RAG significantly outperforms state-of-the-art multi-round and Graph RAG methods, achieving an average performance gain of up to 11% across six datasets while reducing retrieval costs by up to 45%.

Primary Area: foundation or frontier models, including LLMs

Submission Number: 20184