Agentic Deep Graph Reasoning Yields Self-Organizing Knowledge Networks

TL;DR

The paper presents an autonomous, graph-native reasoning framework that iteratively expands a knowledge graph in concert with a reasoning LLM, producing self-organizing, scale-free networks with hubs and interdisciplinary bridge nodes. By embedding reasoning within the graph construction loop and applying extensive network analysis (modularity, betweenness, k-core, path lengths), the work demonstrates open-ended knowledge growth across hundreds of iterations without ontological predefinition. In a materials-design context, agentic reasoning uncovers cross-domain synergies, formulates novel hypotheses (e.g., Bio-Inspired Adaptive Materials for Resilient Ecosystems, BAMES), and yields structured compositional reasoning workflows that outperform baseline approaches. The findings suggest that recursive graph reasoning can serve as a scalable, interpretable, and potentially transformative paradigm for autonomous scientific discovery and knowledge management across domains.

Abstract

We present an agentic, autonomous graph expansion framework that iteratively structures and refines knowledge in situ. Unlike conventional knowledge graph construction methods relying on static extraction or single-pass learning, our approach couples a reasoning-native large language model with a continually updated graph representation. At each step, the system actively generates new concepts and relationships, merges them into a global graph, and formulates subsequent prompts based on its evolving structure. Through this feedback-driven loop, the model organizes information into a scale-free network characterized by hub formation, stable modularity, and bridging nodes that link disparate knowledge clusters. Over hundreds of iterations, new nodes and edges continue to appear without saturating, while centrality measures and shortest path distributions evolve to yield increasingly distributed connectivity. Our analysis reveals emergent patterns, such as the rise of highly connected 'hub' concepts and the shifting influence of 'bridge' nodes, indicating that agentic, self-reinforcing graph construction can yield open-ended, coherent knowledge structures. Applied to materials design problems, we present compositional reasoning experiments by extracting node-specific and synergy-level principles to foster genuinely novel knowledge synthesis, yielding cross-domain ideas that transcend rote summarization and strengthen the framework's potential for open-ended scientific discovery. We discuss other applications in scientific discovery and outline future directions for enhancing scalability and interpretability.

Figures (26)

• Figure 1: Algorithm used for iterative knowledge extraction and graph refinement. At each iteration $i$, the model generates reasoning tokens (blue). From the response, a local graph $\mathcal{G}_{\text{local}}^i$ is extracted (violet) and merged with the global knowledge graph $\mathcal{G}$ (light violet). The evolving graph is stored in multiple formats for visualization and analysis (yellow). Instead of letting the model respond to the task, a follow-up task is generated based on the latest extracted nodes and edges in $\mathcal{G}_{\text{local}}^i$ (green), ensuring iterative refinement (orange), so that the model generates yet more reasoning tokens, and as part of that process, new nodes and edges. The process continues until the stopping condition $i < N$ is met, yielding a final structured knowledge graph $\mathcal{G}$ (orange).
• Figure 2: Knowledge graph $\mathcal{G_1}$ after around 1,000 iterations, under a flexible self-exploration scheme initiated with the prompt Discuss an interesting idea in bio-inspired materials science. We observe the formation of a highly connected graph with multiple hubs and centers.
• Figure 3: Visualizatrion of the knowledge graph Graph 2 after around 500 iterations, under a topic-specific self-exploration scheme initiated with the prompt Describe a way to design impact resistant materials. The graph structure features a complex interwoven but highly connected network with multiple centers.
• Figure 4: Evolution of basic graph properties over recursive iterations, highlighting the emergence of hierarchical structure, hub formation, and adaptive connectivity, for $\mathcal{G_1}$.
• Figure 5: Evolution of key structural properties in the recursively generated knowledge graph $\mathcal{G_1}$: (a) Louvain modularity, showing stable community formation; (b) average shortest path length, highlighting efficient information propagation; and (c) graph diameter, demonstrating bounded hierarchical expansion.