# LogosKG: Hardware-Optimized Scalable and Interpretable Knowledge Graph Retrieval

Source: [LogosKG: Hardware-Optimized Scalable and Interpretable Knowledge Graph Retrieval](<https://www.sota2.com/research/paper/69e85bd77d0172140ec81151>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-23T05:00:05.535Z

Authors: He Cheng, Yifu Wu, Saksham Khatwani, Maya Kruse, Dmitriy Dligach, Timothy A. Miller, Majid Afshar, Yanjun Gao

Published: 2026-04-20T23:32:13.000Z

## Abstract

Knowledge graphs (KGs) are increasingly integrated with large language models (LLMs) to provide structured, verifiable reasoning. A core operation in this integration is multi-hop retrieval, yet existing systems struggle to balance efficiency, scalability, and interpretability. We introduce LogosKG, a novel, hardware-aligned framework that enables scalable and interpretable k-hop retrieval on large KGs by building on symbolic KG formulations and executing traversal as hardware-efficient operations over decomposed subject, object, and relation representations. To scale to billion-edge graphs, LogosKG integrates degree-aware partitioning, cross-graph routing, and on-demand caching. Experiments show substantial efficiency gains over CPU and GPU baselines without loss of retrieval fidelity. With proven performance in KG retrieval, a downstream two-round KG-LLM interaction demonstrates how LogosKG enables large-scale, evidence-grounded analysis of how KG topology, such as hop distribution and connectivity, shapes the alignment between structured biomedical knowledge and LLM diagnostic reasoning, thereby opening the door for next-generation KG-LLM integration. The source code is publicly available at https://github.com/LARK-NLP-Lab/LogosKG, and an online demo is available at https://lark-nlp-lab-logoskg.hf.space/.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Knowledge Graph Retrieval on PKG Hop 3](<https://www.sota2.com/research/sota/knowledge-graph-retrieval-on-pkg-hop-3.md>): Knowledge Graph Retrieval; PKG Hop 3; Query Time (ms): 43.07.

- [Knowledge Graph Retrieval on PKG Hop 4](<https://www.sota2.com/research/sota/knowledge-graph-retrieval-on-pkg-hop-4.md>): Knowledge Graph Retrieval; PKG Hop 4; QT (ms): 77.73.

- [Knowledge Graph Retrieval on PKG Hop 5](<https://www.sota2.com/research/sota/knowledge-graph-retrieval-on-pkg-hop-5.md>): Knowledge Graph Retrieval; PKG Hop 5; QT (ms): 101.05.

- [Knowledge Graph Retrieval on PKG Hop 2](<https://www.sota2.com/research/sota/knowledge-graph-retrieval-on-pkg-hop-2.md>): Knowledge Graph Retrieval; PKG Hop 2; Query Time (ms): 14.4.

- [Knowledge Graph Retrieval on PKG Hop 1](<https://www.sota2.com/research/sota/knowledge-graph-retrieval-on-pkg-hop-1.md>): Knowledge Graph Retrieval; PKG Hop 1; Query Time (ms): 4.76.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.18913>)

[PDF](<https://arxiv.org/pdf/2604.18913.pdf>)