# Multi-Hop Question Answering on MuSiQue (Avg@4)

Source: [Multi-Hop Question Answering on MuSiQue (Avg@4)](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-avg-at-4>)

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

Data updated: 2026-02-26T05:36:32.213Z

Evaluation of model performance on the MuSiQue dataset for Multi-Hop Question Answering, specifically measured using the Avg@4 metric.

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 | Multi-Hop Question Answering |
| Dataset (including any stated split) | MuSiQue |
| Metric | Avg@4 |
| 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

- [Avg@4](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-avg-at-4.md?metric=Avg%404>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Avg@4 | Result date | Report paper |
| --- | --- | --- | --- |
| ASearcher-7B | 32.6 | Not recorded | 6984242fa8b01e76cc8ee83b |
| OWL-8B | 30.4 | Not recorded | 6984242fa8b01e76cc8ee83b |
| WIDESEEK-R1-4B | 28.9 | Not recorded | 6984242fa8b01e76cc8ee83b |
| Search-R1-7B | 27.1 | Not recorded | 6984242fa8b01e76cc8ee83b |
| SingleSeek-R1-4B | 26.5 | Not recorded | 6984242fa8b01e76cc8ee83b |
| AgentFlow-7B | 25.3 | Not recorded | 6984242fa8b01e76cc8ee83b |
| Qwen3-4B | 21.7 | Not recorded | 6984242fa8b01e76cc8ee83b |
| MiroFlow-8B | 21.3 | Not recorded | 6984242fa8b01e76cc8ee83b |
| Qwen3-4B | 19.2 | Not recorded | 6984242fa8b01e76cc8ee83b |

### ASearcher-7B

Avg@4: 32.6. Result ID: 699604dc5698665ec43bcbeb.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b441e209131cea87635.md>)

Recorded evaluation configuration: {"Setting":"Single Agent"}. This record does not assert independent reproduction.

### OWL-8B

Avg@4: 30.4. Result ID: 699604dc5698665ec43bcbef.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795f1e209131cea7d500.md>)

Recorded evaluation configuration: {"Setting":"Multi-Agent System"}. This record does not assert independent reproduction.

### WIDESEEK-R1-4B

Avg@4: 28.9. Result ID: 699604dc5698665ec43bcbec.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Implementation](<https://github.com/RLinf/RLinf/tree/main/examples/wideseek_r1>)

Recorded evaluation configuration: {"Setting":"Multi-Agent System"}. This record does not assert independent reproduction.

### Search-R1-7B

Avg@4: 27.1. Result ID: 699604dc5698665ec43bcbea.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a651e209131cea82e7b.md>)

Recorded evaluation configuration: {"Setting":"Single Agent"}. This record does not assert independent reproduction.

### SingleSeek-R1-4B

Avg@4: 26.5. Result ID: 699604dc5698665ec43bcbe8.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Implementation](<https://github.com/RLinf/RLinf/tree/main/examples/wideseek_r1>)

Recorded evaluation configuration: {"Setting":"Single Agent"}. This record does not assert independent reproduction.

### AgentFlow-7B

Avg@4: 25.3. Result ID: 699604dc5698665ec43bcbee.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

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

Recorded evaluation configuration: {"Setting":"Multi-Agent System"}. This record does not assert independent reproduction.

### Qwen3-4B

Avg@4: 21.7. Result ID: 699604dc5698665ec43bcbed.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Setting":"Multi-Agent System"}. This record does not assert independent reproduction.

### MiroFlow-8B

Avg@4: 21.3. Result ID: 699604dc5698665ec43bcbf0.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

Recorded evaluation configuration: {"Setting":"Multi-Agent System"}. This record does not assert independent reproduction.

### Qwen3-4B

Avg@4: 19.2. Result ID: 699604dc5698665ec43bcbe9.

[Reporting paper](<https://www.sota2.com/research/paper/6984242fa8b01e76cc8ee83b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Setting":"Single Agent"}. This record does not assert independent reproduction.

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