CodaRAG: Connecting the Dots with Associativity Inspired by Complementary Learning

TL;DR

CodaRAG introduces an innovative retrieval framework inspired by CLS, transforming passive lookup into active associative discovery to improve reasoning and factual accuracy in LLMs.

Contribution

It presents a novel three-stage pipeline that enhances evidence retrieval by consolidating knowledge, navigating associative graphs, and eliminating noise.

Findings

Achieves 7-10% higher retrieval recall on GraphRAG-Bench.

Improves generation accuracy by 3-11%.

Demonstrates robustness in factual, reasoning, and creative tasks.

Abstract

Large Language Models (LLMs) struggle with knowledge-intensive tasks due to hallucinations and fragmented reasoning over dispersed information. While Retrieval-Augmented Generation (RAG) grounds generation in external sources, existing methods often treat evidence as isolated units, failing to reconstruct the logical chains that connect these dots. Inspired by Complementary Learning Systems (CLS), we propose CodaRAG, a framework that evolves retrieval from passive lookup into active associative discovery. CodaRAG operates via a three-stage pipeline: (1) Knowledge Consolidation to unify fragmented extractions into a stable memory substrate; (2) Associative Navigation to traverse the graph via multi-dimensional pathways-semantic, contextualized, and functional-explicitly recovering dispersed evidence chains; and (3) Interference Elimination to prune hyper-associative noise, ensuring a coherent, high-precision reasoning context. On GraphRAG-Bench, CodaRAG achieves absolute gains of 7-10% in retrieval recall and 3-11% in generation accuracy. These results demonstrate CodaRAG's superior ability to systematically robustify associative evidence retrieval for factual, reasoning, and creative tasks.