# Knowledge Graph Retrieval on PKG Hop 4 Source: [Knowledge Graph Retrieval on PKG Hop 4]() Collection index: [llms.txt]() Data updated: 2026-04-23T07:38:06.896Z Evaluation of model performance on the PKG Hop 4 dataset for Knowledge Graph Retrieval, measured using metrics like QT (ms) and TR (%). 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. | Scope | Recorded value | | --- | --- | | Task | Knowledge Graph Retrieval | | Dataset (including any stated split) | PKG Hop 4 | | Metric | QT (ms) | | Unit | TIME | | Direction | Lower is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [QT (ms)](): Lower is better; unit: TIME. - [TR (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | QT (ms) | Result date | Report paper | | --- | --- | --- | --- | | LogosKG (Torch-GPU) | 77.73 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Numba) | 140.07 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (SciPy) | 289.61 | Not recorded | 69e85bd77d0172140ec81151 | | SNAP | 378.81 | Not recorded | 69e85bd77d0172140ec81151 | | GraphBLAS | 528.07 | Not recorded | 69e85bd77d0172140ec81151 | | NetworkX | 621.95 | Not recorded | 69e85bd77d0172140ec81151 | | PyG | 646.96 | Not recorded | 69e85bd77d0172140ec81151 | | igraph | 837.12 | Not recorded | 69e85bd77d0172140ec81151 | | DGL | 1121.7 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Large-Numba) | 1411.72 | Not recorded | 69e85bd77d0172140ec81151 | | cuGraph | 1504.82 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Large-Torch-GPU) | 1634.93 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Large-SciPy) | 1660.91 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Torch-CPU) | 1803.18 | Not recorded | 69e85bd77d0172140ec81151 | | graph-tool | 2396.95 | Not recorded | 69e85bd77d0172140ec81151 | | LogosKG (Large-Torch-CPU) | 4836.16 | Not recorded | 69e85bd77d0172140ec81151 | | Neo4j | 8000 | Not recorded | 69e85bd77d0172140ec81151 | ### LogosKG (Torch-GPU) QT (ms): 77.73. Result ID: 69e9cc4047f34ce924d0a5c9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Torch","Hardware":"GPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### LogosKG (Numba) QT (ms): 140.07. Result ID: 69e9cc4047f34ce924d0a5c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Numba","Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### LogosKG (SciPy) QT (ms): 289.61. Result ID: 69e9cc4047f34ce924d0a5c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"SciPy","Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### SNAP QT (ms): 378.81. Result ID: 69e9cc4047f34ce924d0a5c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### GraphBLAS QT (ms): 528.07. Result ID: 69e9cc4047f34ce924d0a5c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### NetworkX QT (ms): 621.95. Result ID: 69e9cc4047f34ce924d0a5bd. [Reporting paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### PyG QT (ms): 646.96. Result ID: 69e9cc4047f34ce924d0a5c5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hardware":"GPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### igraph QT (ms): 837.12. Result ID: 69e9cc4047f34ce924d0a5be. [Reporting paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### DGL QT (ms): 1121.7. Result ID: 69e9cc4047f34ce924d0a5c4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Hardware":"GPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### LogosKG (Large-Numba) QT (ms): 1411.72. Result ID: 69e9cc4047f34ce924d0a5ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Numba","Hardware":"CPU","Scale Variant":"Large"}. This record does not assert independent reproduction. ### cuGraph QT (ms): 1504.82. Result ID: 69e9cc4047f34ce924d0a5c3. [Reporting paper]() Recorded evaluation configuration: {"Hardware":"GPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### LogosKG (Large-Torch-GPU) QT (ms): 1634.93. Result ID: 69e9cc4047f34ce924d0a5cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Torch","Hardware":"GPU","Scale Variant":"Large"}. This record does not assert independent reproduction. ### LogosKG (Large-SciPy) QT (ms): 1660.91. Result ID: 69e9cc4047f34ce924d0a5cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"SciPy","Hardware":"CPU","Scale Variant":"Large"}. This record does not assert independent reproduction. ### LogosKG (Torch-CPU) QT (ms): 1803.18. Result ID: 69e9cc4047f34ce924d0a5c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Torch","Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### graph-tool QT (ms): 2396.95. Result ID: 69e9cc4047f34ce924d0a5bf. [Reporting paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. ### LogosKG (Large-Torch-CPU) QT (ms): 4836.16. Result ID: 69e9cc4047f34ce924d0a5cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backend":"Torch","Hardware":"CPU","Scale Variant":"Large"}. This record does not assert independent reproduction. ### Neo4j QT (ms): 8000. Result ID: 69e9cc4047f34ce924d0a5c2. [Reporting paper]() Recorded evaluation configuration: {"Hardware":"CPU","Scale Variant":"Standard"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.