# Open-domain Question Answering on NQ 3.5K (test)

Source: [Open-domain Question Answering on NQ 3.5K (test)](<https://www.sota2.com/research/sota/open-domain-question-answering-on-nq-3-5k-test>)

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

Data updated: 2026-02-26T05:41:40.140Z

Evaluation of Open-domain Question Answering performance on the NQ 3.5K test split, measured using EM and F1 scores.

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) | NQ 3.5K (test) |
| Metric | EM |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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](<https://www.sota2.com/research/sota/open-domain-question-answering-on-nq-3-5k-test.md?metric=EM>): Higher is better; unit: PERCENT.

- [F1](<https://www.sota2.com/research/sota/open-domain-question-answering-on-nq-3-5k-test.md?metric=F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EM | Result date | Report paper |
| --- | --- | --- | --- |
| QNorm | 0.261 | Not recorded | 698c0cfcc97a0b3e3b5e10e2 |
| Learn | 0.25 | Not recorded | 698c0cfcc97a0b3e3b5e10e2 |
| Dot | 0.245 | Not recorded | 698c0cfcc97a0b3e3b5e10e2 |
| Cosine | 0.23 | Not recorded | 698c0cfcc97a0b3e3b5e10e2 |
| DNorm | 0.228 | Not recorded | 698c0cfcc97a0b3e3b5e10e2 |

### QNorm

EM: 0.261. Result ID: 6998d0575698665ec433f887.

[Reporting paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

Recorded evaluation configuration: {"Retriever":"Contriever (fine-tuned on MS MARCO 80K)","Reader":"Flan-T5-Large","Similarity":"QNorm"}. This record does not assert independent reproduction.

### Learn

EM: 0.25. Result ID: 6998d0575698665ec433f889.

[Reporting paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

Recorded evaluation configuration: {"Retriever":"Contriever (fine-tuned on MS MARCO 80K)","Reader":"Flan-T5-Large","Similarity":"Learn (learnable normalization)"}. This record does not assert independent reproduction.

### Dot

EM: 0.245. Result ID: 6998d0575698665ec433f886.

[Reporting paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

Recorded evaluation configuration: {"Retriever":"Contriever (fine-tuned on MS MARCO 80K)","Reader":"Flan-T5-Large","Similarity":"Dot"}. This record does not assert independent reproduction.

### Cosine

EM: 0.23. Result ID: 6998d0575698665ec433f885.

[Reporting paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

Recorded evaluation configuration: {"Retriever":"Contriever (fine-tuned on MS MARCO 80K)","Reader":"Flan-T5-Large","Similarity":"Cosine"}. This record does not assert independent reproduction.

### DNorm

EM: 0.228. Result ID: 6998d0575698665ec433f888.

[Reporting paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698c0cfcc97a0b3e3b5e10e2.md>)

Recorded evaluation configuration: {"Retriever":"Contriever (fine-tuned on MS MARCO 80K)","Reader":"Flan-T5-Large","Similarity":"DNorm"}. This record does not assert independent reproduction.

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