# Multi-hop Question Answering on MuSiQue 23 (test)

Source: [Multi-hop Question Answering on MuSiQue 23 (test)](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test>)

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

Data updated: 2026-05-22T07:58:38.201Z

Evaluation of model performance on the test split of the MuSiQue 23 dataset for Multi-hop Question Answering, utilizing specific metrics such as p-hat, alpha-hat, beta-hat, RelLat-star, and RelLat.

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 23 (test) |
| Metric | p-hat |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [p-hat](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test.md?metric=p-hat>): Higher is better; unit: SCALAR.

- [alpha-hat](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test.md?metric=alpha-hat>): Higher is better; unit: SCALAR.

- [beta-hat](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test.md?metric=beta-hat>): Higher is better; unit: SCALAR.

- [RelLat-star](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test.md?metric=RelLat-star>): Higher is better; unit: SCALAR.

- [RelLat](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-musique-23-test.md?metric=RelLat>): Higher is better; unit: SCALAR.

## Recorded results

| Method | p-hat | Result date | Report paper |
| --- | --- | --- | --- |
| SPECHOP | 0.5 | Not recorded | 6a0fe326f984739ed6739e79 |
| SPECHOP | 0.49 | Not recorded | 6a0fe326f984739ed6739e79 |
| SPECHOP | 0.42 | Not recorded | 6a0fe326f984739ed6739e79 |
| SPECHOP | 0.41 | Not recorded | 6a0fe326f984739ed6739e79 |
| SPECHOP | 0.34 | Not recorded | 6a0fe326f984739ed6739e79 |
| SPECHOP | 0.33 | Not recorded | 6a0fe326f984739ed6739e79 |

### SPECHOP

p-hat: 0.5. Result ID: 6a100caafbc373592c08cc08.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"Web Search","Speculator (S)":"GPT-4o"}. This record does not assert independent reproduction.

### SPECHOP

p-hat: 0.49. Result ID: 6a100caafbc373592c08cc07.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"Web Search","Speculator (S)":"GPT-4o mini"}. This record does not assert independent reproduction.

### SPECHOP

p-hat: 0.42. Result ID: 6a100caafbc373592c08cc03.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"E5 Retrieval","Speculator (S)":"Llama 3.1 8B"}. This record does not assert independent reproduction.

### SPECHOP

p-hat: 0.41. Result ID: 6a100caafbc373592c08cc05.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"Web Search","Speculator (S)":"Llama 3.1 8B"}. This record does not assert independent reproduction.

### SPECHOP

p-hat: 0.34. Result ID: 6a100caafbc373592c08cc04.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"E5 Retrieval","Speculator (S)":"Qwen 3 8B"}. This record does not assert independent reproduction.

### SPECHOP

p-hat: 0.33. Result ID: 6a100caafbc373592c08cc06.

[Reporting paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0fe326f984739ed6739e79.md>)

Recorded evaluation configuration: {"Target Tool (T)":"Web Search","Speculator (S)":"Qwen 3 8B"}. This record does not assert independent reproduction.

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