# Answering EPFO queries on FB15k (test)

Source: [Answering EPFO queries on FB15k (test)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test>)

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

Data updated: 2026-02-26T05:50:31.163Z

Evaluation of model performance on the test split of the FB15k dataset specifically for answering EPFO-related knowledge graph queries, measured using detailed Mean Reciprocal Rank (MRR) path-based metrics.

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 | Answering EPFO queries |
| Dataset (including any stated split) | FB15k (test) |
| Metric | MRR (1-path) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [MRR (1-path)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(1-path)>): Higher is better; unit: SCALAR.

- [MRR (2-path)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(2-path)>): Higher is better; unit: SCALAR.

- [MRR (3-path)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(3-path)>): Higher is better; unit: SCALAR.

- [MRR (2-inverse)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(2-inverse)>): Higher is better; unit: SCALAR.

- [MRR (3-inverse)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(3-inverse)>): Higher is better; unit: SCALAR.

- [MRR (path-inverse)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(path-inverse)>): Higher is better; unit: SCALAR.

- [MRR (inverse-path)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(inverse-path)>): Higher is better; unit: SCALAR.

- [MRR (2u DNF)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(2u%20DNF)>): Higher is better; unit: SCALAR.

- [MRR (2u DM)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(2u%20DM)>): Higher is better; unit: SCALAR.

- [MRR (up DNF)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(up%20DNF)>): Higher is better; unit: SCALAR.

- [MRR (up DM)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(up%20DM)>): Higher is better; unit: SCALAR.

- [MRR (Average)](<https://www.sota2.com/research/sota/answering-epfo-queries-on-fb15k-test.md?metric=MRR%20(Average)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | MRR (1-path) | Result date | Report paper |
| --- | --- | --- | --- |
| Q2B | 0.68 | Not recorded | 68fa688a1e209131cea1cc9c |
| BETAE | 0.651 | Not recorded | 68fa688a1e209131cea1cc9c |
| GQE | 0.546 | Not recorded | 68fa688a1e209131cea1cc9c |

### Q2B

MRR (1-path): 0.68. Result ID: 698d76648b517eb43321af52.

[Reporting paper](<https://www.sota2.com/research/paper/68fa688a1e209131cea1cc9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67221e209131cea148ab.md>)

[Implementation](<http://snap.stanford.edu/query2box>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### BETAE

MRR (1-path): 0.651. Result ID: 698d76648b517eb43321af51.

[Reporting paper](<https://www.sota2.com/research/paper/68fa688a1e209131cea1cc9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa688a1e209131cea1cc9c.md>)

[Implementation](<http://snap.stanford.edu/betae>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GQE

MRR (1-path): 0.546. Result ID: 698d76648b517eb43321af53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa688a1e209131cea1cc9c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66911e209131cea10f0a.md>)

[Implementation](<https://github.com/williamleif/graphqembed>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.