Beyond Chunks and Graphs: Retrieval-Augmented Generation through Triplet-Driven Thinking
Shengbo Gong, Xianfeng Tang, Qi He, Carl Yang, Jin Yu Wei · 2026
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 multiround and Graph RAG methods, achieving an average performance gain of up to 11% across six datasets while reducing retrieval costs by up to 45%.Our code is available at https:// anonymous.4open.science/r/T2RAG-DF75. 1 26282 (a) Multi-round RAG (b) Graph RAG (c) Triplet-driven Thinking RAG (Ours)Question: Which film has the director born earlier, God's Gift To Women or Aldri Annet Enn Brak?I need to find the birth years of the directors of both films: Michael Curtiz for "God's Gift to Women" and Edith Carlmar for "Aldri annet enn bråk....