# Open-domain Question Answering on Natural Questions Open (dev) Source: [Open-domain Question Answering on Natural Questions Open (dev)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:41:32.628Z Evaluation of model performance on the development split of the Natural Questions Open dataset for open-domain question answering, measured using Exact Match (EM). 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 | Open-domain Question Answering | | Dataset (including any stated split) | Natural Questions Open (dev) | | Metric | EM | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 9 of 9 | | 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 - [EM](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM | Result date | Report paper | | --- | --- | --- | --- | | SPARTA | 36.8 | Not recorded | 68fa68631e209131cea1bb2d | | Recurrent Graph-based Retrieval | 31.7 | Not recorded | 68fa670a1e209131cea1406e | | Graph-retriever | 31.7 | Not recorded | 68fa68631e209131cea1bb2d | | ORQA | 31.3 | Not recorded | 68fa670a1e209131cea1406e | | ORQA | 31.3 | Not recorded | 68fa68631e209131cea1bb2d | | Hard EM | 28.8 | Not recorded | 68fa670a1e209131cea1406e | | Hard EM | 28.8 | Not recorded | 68fa68631e209131cea1bb2d | | BERT + BM 25 | 24.8 | Not recorded | 68fa670a1e209131cea1406e | | BERT + BM25 | 24.8 | Not recorded | 68fa68631e209131cea1bb2d | ### SPARTA EM: 36.8. Result ID: 698d6c498b517eb433216578. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Ranker Backbone":"bert-base-uncased","Reader Backbone":"SpanBERT"}. This record does not assert independent reproduction. ### Recurrent Graph-based Retrieval EM: 31.7. Result ID: 698d169d8b517eb4331eea79. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Graph-retriever EM: 31.7. Result ID: 698d6c498b517eb433216577. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ORQA EM: 31.3. Result ID: 698d169d8b517eb4331eea76. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ORQA EM: 31.3. Result ID: 698d6c498b517eb433216576. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Hard EM EM: 28.8. Result ID: 698d169d8b517eb4331eea77. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Hard EM EM: 28.8. Result ID: 698d6c498b517eb433216575. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT + BM 25 EM: 24.8. Result ID: 698d169d8b517eb4331eea78. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT + BM25 EM: 24.8. Result ID: 698d6c498b517eb433216574. [Reporting paper]() [Method origin paper]() 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.