# Open-domain Question Answering on NQ-Open OOD (test)

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

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

Data updated: 2026-05-15T07:38:03.967Z

Evaluation of Open-domain Question Answering performance on the NQ-Open Out-of-Distribution (OOD) test set, measured using Exact Match (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-Open OOD (test) |
| Metric | Exact Match (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

- [Exact Match (EM)](<https://www.sota2.com/research/sota/open-domain-question-answering-on-nq-open-ood-test.md?metric=Exact%20Match%20(EM)>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/open-domain-question-answering-on-nq-open-ood-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Exact Match (EM) | Result date | Report paper |
| --- | --- | --- | --- |
| SkillFlow | 82.81 | Not recorded | 6a06a89bdbc836b87d82a078 |
| FlowSteer | 80.47 | Not recorded | 6a06a89bdbc836b87d82a078 |
| AgentFlow | 78.91 | Not recorded | 6a06a89bdbc836b87d82a078 |
| SkillRL | 78.91 | Not recorded | 6a06a89bdbc836b87d82a078 |
| AFlow | 75.79 | Not recorded | 6a06a89bdbc836b87d82a078 |
| v4-flash | 37.5 | Not recorded | 6a06a89bdbc836b87d82a078 |
| Qwen3.5 | 23.44 | Not recorded | 6a06a89bdbc836b87d82a078 |
| Qwen3.5 | 21.88 | Not recorded | 6a06a89bdbc836b87d82a078 |
| Qwen3.5 | 21.88 | Not recorded | 6a06a89bdbc836b87d82a078 |

### SkillFlow

Exact Match (EM): 82.81. Result ID: 6a06cd5147f34ce924ee1726.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

Recorded evaluation configuration: {"Method Type":"Ours","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### FlowSteer

Exact Match (EM): 80.47. Result ID: 6a06cd5147f34ce924ee1724.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698181e5a8b01e76cc8e28fa.md>)

Recorded evaluation configuration: {"Method Type":"Agent+RL","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### AgentFlow

Exact Match (EM): 78.91. Result ID: 6a06cd5147f34ce924ee1723.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bf01e209131cea8b6fa.md>)

Recorded evaluation configuration: {"Method Type":"Agent+RL","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### SkillRL

Exact Match (EM): 78.91. Result ID: 6a06cd5147f34ce924ee1725.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698abbd1c97a0b3e3b5de1d0.md>)

[Implementation](<https://github.com/aiming-lab/SkillRL>)

Recorded evaluation configuration: {"Method Type":"Agent+RL","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### AFlow

Exact Match (EM): 75.79. Result ID: 6a06cd5147f34ce924ee1722.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78111e209131cea769ef.md>)

Recorded evaluation configuration: {"Method Type":"Search-based workflow","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### v4-flash

Exact Match (EM): 37.5. Result ID: 6a06cd5147f34ce924ee171f.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

Recorded evaluation configuration: {"Method Type":"Baseline","Backbone Model":"v4-flash"}. This record does not assert independent reproduction.

### Qwen3.5

Exact Match (EM): 23.44. Result ID: 6a06cd5147f34ce924ee1720.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

Recorded evaluation configuration: {"Method Type":"SFT","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### Qwen3.5

Exact Match (EM): 21.88. Result ID: 6a06cd5147f34ce924ee171e.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

Recorded evaluation configuration: {"Method Type":"Baseline","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

### Qwen3.5

Exact Match (EM): 21.88. Result ID: 6a06cd5147f34ce924ee1721.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a89bdbc836b87d82a078.md>)

Recorded evaluation configuration: {"Method Type":"GRPO","Backbone Model":"Qwen3.5-9B"}. This record does not assert independent reproduction.

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