Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GroupRAG: Cognitively Inspired Group-Aware Retrieval and Reasoning via Knowledge-Driven Problem Structuring

Xinyi Duan, Yuanrong Tang, Jiangtao Gong

Abstract

The performance of language models is commonly limited by insufficient knowledge and constrained reasoning. Prior approaches such as Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT) address these issues by incorporating external knowledge or enforcing linear reasoning chains, but often degrade in real-world settings. Inspired by cognitive science, which characterizes human problem solving as search over structured problem spaces rather than single inference chains, we argue that inadequate awareness of problem structure is a key overlooked limitation. We propose GroupRAG, a cognitively inspired, group-aware retrieval and reasoning framework based on knowledge-driven keypoint grouping. GroupRAG identifies latent structural groups within a problem and performs retrieval and reasoning from multiple conceptual starting points, enabling fine-grained interaction between the two processes. Experiments on MedQA show that GroupRAG outperforms representative RAG- and CoT-based baselines. These results suggest that explicitly modeling problem structure, as inspired by human cognition, is a promising direction for robust retrieval-augmented reasoning.

GroupRAG: Cognitively Inspired Group-Aware Retrieval and Reasoning via Knowledge-Driven Problem Structuring

Abstract

The performance of language models is commonly limited by insufficient knowledge and constrained reasoning. Prior approaches such as Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT) address these issues by incorporating external knowledge or enforcing linear reasoning chains, but often degrade in real-world settings. Inspired by cognitive science, which characterizes human problem solving as search over structured problem spaces rather than single inference chains, we argue that inadequate awareness of problem structure is a key overlooked limitation. We propose GroupRAG, a cognitively inspired, group-aware retrieval and reasoning framework based on knowledge-driven keypoint grouping. GroupRAG identifies latent structural groups within a problem and performs retrieval and reasoning from multiple conceptual starting points, enabling fine-grained interaction between the two processes. Experiments on MedQA show that GroupRAG outperforms representative RAG- and CoT-based baselines. These results suggest that explicitly modeling problem structure, as inspired by human cognition, is a promising direction for robust retrieval-augmented reasoning.

Paper Structure

This paper contains 30 sections, 8 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Chain-of-Thought
  4. Retrieval-Augmented Generation
  5. Reasoning in Cognitive Science
  6. Method
  7. Group-Aware Retrieval and Reasoning
  8. System Design
  9. Modular Pipeline and Stage-wise Training
  10. Global Reasoning Optimization
  11. Weighted Inference F-score (WIF).
  12. Policy Optimization.
  13. Retrieval-Augmented Generation in GroupRAG
  14. Experiments
  15. Dataset and Model
  16. ...and 15 more sections

Figures (3)

  • Figure 1: Reasoning Paradigms Comparison. Traditional CoT follows Linear/Divergent paths on unstructured sequences. GroupRAG transforms monolithic inputs into a structured problem space, employing a Convergent net via keypoint grouping for real-world alignment.
  • Figure 2: Illustration of the GroupRAG reasoning process on a clinical case, featuring keypoint extraction, knowledge-driven grouping, local and global reasoning, and answer alignment.
  • Figure 3: An Abstract Overview of GroupRAG.