CoG (Cognition on Graph), a cognitive-inspired, training-free framework for adaptive knowledge exploration, which establishes deep bidirectional synergy between structured graph and unstructured text, where entities extracted from text dynamically guide graph exploration to bridge knowledge gaps.
Abstract
Retrieval-Augmented Generation (RAG) has empowered Large Language Models (LLMs) to tackle knowledge-intensive tasks. However, navigating global, heterogeneous knowledge bases (large-scale knowledge graphs and text corpora) for complex reasoning remains a challenge. Existing methods typically employ reactive, graph-driven exploration strategies, which blindly follow graph topology without adapting to the question context or evolving exploration progress, and lack deep bidirectional synergy between graph and text. To address these limitations, we propose CoG (Cognition on Graph), a cognitive-inspired, training-free framework for adaptive knowledge exploration. Drawing inspiration from human problem-solving, CoG performs a continuous plan-explore-reflect cycle, where it proactively formulates investigation plans, performs dual-source retrieval, and dynamically reflects on progress to adjust strategies. Crucially, it establishes deep bidirectional synergy between structured graph and unstructured text, where entities extracted from text dynamically guide graph exploration to bridge knowledge gaps. Extensive experiments on seven multi-hop QA benchmarks demonstrate that CoG significantly outperforms state-of-the-art methods while achieving superior exploration efficiency. Our code and datasets are available at https://github.com/zhougengxian/CoG.
Retrieval-augmented generation (RAG) frequently integrates knowledge graphs (KGs) for complex multi-hop question answering. However, their performance is increasingly bottlenecked by the knowledge structuring and reasoning processes. (i) Severe semantic loss: Distilling continuous texts into discrete symbols strips awa...
Yu-Feng Li, Zai-Jing Li, Fu-Chen Zhang et al.· Conference on Empirical Meth...· 0 citations
A novel framework KD-GAG is proposed that efficiently constructs KGs by distilling teacher LLMs' semantics and reasoning into a smaller student LLM and a preference-based subgraph pruning method to optimize the retrieval process.
Long Zhao, Yin Xu, Yan-Yan Wang et al.· Neural Networks· 0 citations
These results demonstrate that collaborative modeling of instance-level and batch-level structural context can effectively enhance structure-aware entity representation and improve fine-grained entity prediction.
RACER employs a semantic-aware action pruning and teacher-guided reinforcement learning mechanism to efficiently extract high-quality reasoning pathways from large-scale KGs, and introduces a cross-task accumulated shared memory graph paired with an attention-driven multi-path knowledge refinement module.
Yuwei Lou, Hao Hu, Yu-Zhou Jiang et al.· 1 citation
Large language models (LLMs) have demonstrated strong capabilities in question answering, yet they still frequently suffer from hallucinations on knowledge-intensive tasks. Knowledge graphs (KGs) provide LLMs with structured, interpretable, and updatable factual grounding, making them a promising external knowledge sou...
Yang Hong, Ya-Jun Yang, Xin Wang et al.· 0 citations