semantic graph modeling

Existing graph database (GDB) models lack hardware–software co-design, resulting in low storage density and inefficient graph traversal—hindering high-performance semantic reasoning required by symbolic AI and retrieval-augmented generation (RAG). To address this, we propose Views, a hardware-accelerated GDB model that redefines the graph data structure through compact symbolic encoding, memory-aligned layout, and hardware-friendly traversal primitives—preserving semantic equivalence while significantly improving storage density and random-access efficiency. Experimental evaluation on representative cognitive modeling and RAG knowledge retrieval tasks demonstrates that Views achieves 2.3×–5.1× higher throughput and reduces latency by 62%–79% compared to state-of-the-art GDBs. Moreover, Views enables scalable symbolic knowledge representation and reasoning, bridging the gap between expressive graph semantics and hardware-efficient execution.

This paper addresses the challenge of cross-scale modeling in data spaces by proposing a Multi-level Graph (MLG) structure that enables multi-granularity data abstraction—from local to global. We formalize topological contraction and expansion operations to establish an incremental, invertible graph transformation algebra, unifying the representation of both structured and unstructured data. Unlike conventional single-layer graph models, our approach achieves the first semantic-preserving hierarchical compression and expansion of data. Experiments on a real-world dream report dataset demonstrate a 42% improvement in cross-granularity exploratory efficiency and significantly enhanced semantic coherence. This work introduces a scalable, interpretable, and dynamically evolvable foundational representation paradigm for data spaces.

Large language models (LLMs) struggle to effectively comprehend graph-structured data due to their inherent sequence-based architecture and lack of native graph-aware representations. Method: This paper introduces *graph laws*—statistically derived, topologically parameterized features that are interpretable as natural language descriptions—establishing a novel paradigm for representing graphs as LLM-compatible inputs. We systematically construct a multi-dimensional graph law framework spanning macro/micro scales, low/high orders, and static/dynamic properties, integrating graph-theoretic analysis, multi-scale observational modeling, and natural language alignment techniques, while establishing semantic mappings to downstream graph tasks and retrieval-augmented generation (RAG) scenarios. Results: Experiments demonstrate that graph laws substantially mitigate LLM hallucination, overcome context-length limitations, and enable end-to-end graph reasoning. The approach achieves strong generalization across diverse domains, including molecular design, recommender systems, and protein structure modeling.

Existing graph representations—such as adjacency matrices—are incompatible with the text-processing paradigm of large language models (LLMs). To address this, we propose a reversible, locally structure-preserving mapping from graphs to instruction sequences: graphs are encoded into compact, deterministic instruction strings via adjacency matrix decomposition, yielding a parseable textual representation; a custom reversible parser enables lossless reconstruction. This work establishes the first bridge between graph algebra and LLM-native textual processing, simultaneously ensuring structural fidelity and sequence conciseness. Experiments demonstrate that our representation significantly improves LLM performance on graph modeling tasks—including graph classification and link prediction—validating both its effectiveness and generalizability across diverse graph domains.

This study addresses the challenge of unifying attribute graph models and SQL querying within relational databases. The authors propose reinterpreting SQL foreign key semantics as reference keys, enabling natural modeling of labeled property graphs directly on standard relational table structures. They further extend SQL to support efficient graph data insertion and complex pattern matching. This approach achieves deep integration of relational and graph models within a single system without requiring an additional storage engine. Experimental results demonstrate that the proposed method effectively enables graph structure construction and advanced graph querying capabilities, significantly enhancing relational databases’ support for graph-oriented operations.

Existing graph databases lack effective support for the tree-shaped substructures commonly found in property graphs. This work addresses this limitation by treating such tree substructures as first-class citizens and proposes a systematic management framework encompassing modeling, indexing, and query optimization. Drawing inspiration from XML structural indexing techniques, the approach enables efficient path queries within a relational graph database backend. Experimental evaluation demonstrates that the proposed method significantly improves path query performance, thereby validating the potential of structural indexing to enhance graph data management.

This work addresses the limitations of traditional knowledge graph construction approaches, wherein structural decisions are hard-coded into rigid pipelines, resulting in tight coupling between schema and construction process and hindering support for ontology-level tasks. To overcome this, the authors propose an ontology-oriented construction framework featuring a novel intrinsic-relational routing mechanism. This mechanism dynamically assigns attributes to corresponding schema modules through iterative attribute classification, enabling a declarative and reusable decoupled design. The pipeline integrates rule-based cleaning, tool-augmented large language model–assisted annotation, and human review. Evaluated on Wikidata (January 2026), the resulting graph comprises 34 million nodes and 61.2 million edges, achieving 93.3% schema coverage and 98.0% module assignment accuracy, effectively supporting five ontology-level applications.

Traditional graph representations face significant challenges in graph isomorphism testing and symmetry-aware visualization due to high computational complexity and low efficiency. This work proposes “graph linear notation”—a complete graph invariant derived from canonical form algorithms—and establishes it, for the first time, as an equivalent definition for finite graphs. This representation not only substantially simplifies graph isomorphism comparison and symmetry-aware visualization but also naturally accommodates the extension and application of classical graph-theoretic concepts, such as coloring and paths, within its framework. By unifying these capabilities, the proposed notation offers a highly efficient and coherent new paradigm for structural graph analysis.