# Multi-Hop Question Answering on Multi-Hop QA Aggregate

Source: [Multi-Hop Question Answering on Multi-Hop QA Aggregate](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-multi-hop-qa-aggregate>)

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

Data updated: 2026-03-19T08:48:53.101Z

Evaluation of model performance on the Multi-Hop QA Aggregate dataset for multi-hop question answering, reporting average performance and relative change metrics.

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) | Multi-Hop QA Aggregate |
| Metric | Average Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 30 of 30 |
| 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

- [Average Score](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-multi-hop-qa-aggregate.md?metric=Average%20Score>): Higher is better; unit: SCALAR.

- [Relative Change (%)](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-multi-hop-qa-aggregate.md?metric=Relative%20Change%20(%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Score | Result date | Report paper |
| --- | --- | --- | --- |
| Tree-GRPO | 45.3 | Not recorded | 68fa7c101e209131cea8c349 |
| GSPO | 44.2 | Not recorded | 68fa7c101e209131cea8c349 |
| GRPO | 41.8 | Not recorded | 68fa7c101e209131cea8c349 |
| Tree-GRPO | 37.8 | Not recorded | 68fa7c101e209131cea8c349 |
| Tree-GRPO | 36.8 | Not recorded | 68fa7c101e209131cea8c349 |
| Tree-GRPO | 36.8 | Not recorded | 68fa7c101e209131cea8c349 |
| GRPO | 36.4 | Not recorded | 68fa7c101e209131cea8c349 |
| GSPO | 35.4 | Not recorded | 68fa7c101e209131cea8c349 |
| GSPO | 33.5 | Not recorded | 68fa7c101e209131cea8c349 |
| ReAct | 32.8 | Not recorded | 68fa7c101e209131cea8c349 |
| GRPO | 31.8 | Not recorded | 68fa7c101e209131cea8c349 |
| GRPO | 26.7 | Not recorded | 68fa7c101e209131cea8c349 |
| Search-o1 | 23.6 | Not recorded | 68fa7c101e209131cea8c349 |
| GSPO | 23.3 | Not recorded | 68fa7c101e209131cea8c349 |
| ReAct | 23.3 | Not recorded | 68fa7c101e209131cea8c349 |
| Search-o1 | 20.3 | Not recorded | 68fa7c101e209131cea8c349 |
| Tree-GRPO | 19.1 | Not recorded | 68fa7c101e209131cea8c349 |
| Search-o1 | 17.9 | Not recorded | 68fa7c101e209131cea8c349 |
| Direct Inference | 16.9 | Not recorded | 68fa7c101e209131cea8c349 |
| ReAct | 15.8 | Not recorded | 68fa7c101e209131cea8c349 |
| Direct Inference | 14.6 | Not recorded | 68fa7c101e209131cea8c349 |
| Search-o1 | 14.1 | Not recorded | 68fa7c101e209131cea8c349 |
| ReAct | 13.3 | Not recorded | 68fa7c101e209131cea8c349 |
| GSPO | 11.7 | Not recorded | 68fa7c101e209131cea8c349 |
| GRPO | 11.3 | Not recorded | 68fa7c101e209131cea8c349 |
| Direct Inference | 10.9 | Not recorded | 68fa7c101e209131cea8c349 |
| Search-o1 | 10 | Not recorded | 68fa7c101e209131cea8c349 |
| Direct Inference | 8 | Not recorded | 68fa7c101e209131cea8c349 |
| ReAct | 7.1 | Not recorded | 68fa7c101e209131cea8c349 |
| Direct Inference | 5.2 | Not recorded | 68fa7c101e209131cea8c349 |

### Tree-GRPO

Average Score: 45.3. Result ID: 69bbb87247f34ce924a5a85f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### GSPO

Average Score: 44.2. Result ID: 69bbb87247f34ce924a5a85e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### GRPO

Average Score: 41.8. Result ID: 69bbb87247f34ce924a5a85d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### Tree-GRPO

Average Score: 37.8. Result ID: 69bbb87247f34ce924a5a859.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### Tree-GRPO

Average Score: 36.8. Result ID: 69bbb87247f34ce924a5a84d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### Tree-GRPO

Average Score: 36.8. Result ID: 69bbb87247f34ce924a5a853.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### GRPO

Average Score: 36.4. Result ID: 69bbb87247f34ce924a5a857.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### GSPO

Average Score: 35.4. Result ID: 69bbb87247f34ce924a5a858.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### GSPO

Average Score: 33.5. Result ID: 69bbb87247f34ce924a5a84c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### ReAct

Average Score: 32.8. Result ID: 69bbb87247f34ce924a5a85c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### GRPO

Average Score: 31.8. Result ID: 69bbb87247f34ce924a5a84b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### GRPO

Average Score: 26.7. Result ID: 69bbb87247f34ce924a5a851.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### Search-o1

Average Score: 23.6. Result ID: 69bbb87247f34ce924a5a85b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76911e209131cea6dca9.md>)

[Implementation](<https://github.com/sunnynexus/Search-o1>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### GSPO

Average Score: 23.3. Result ID: 69bbb87247f34ce924a5a852.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### ReAct

Average Score: 23.3. Result ID: 69bbb87247f34ce924a5a856.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### Search-o1

Average Score: 20.3. Result ID: 69bbb87247f34ce924a5a855.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76911e209131cea6dca9.md>)

[Implementation](<https://github.com/sunnynexus/Search-o1>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### Tree-GRPO

Average Score: 19.1. Result ID: 69bbb87247f34ce924a5a847.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

### Search-o1

Average Score: 17.9. Result ID: 69bbb87247f34ce924a5a849.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76911e209131cea6dca9.md>)

[Implementation](<https://github.com/sunnynexus/Search-o1>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### Direct Inference

Average Score: 16.9. Result ID: 69bbb87247f34ce924a5a85a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-14b"}. This record does not assert independent reproduction.

### ReAct

Average Score: 15.8. Result ID: 69bbb87247f34ce924a5a84a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### Direct Inference

Average Score: 14.6. Result ID: 69bbb87247f34ce924a5a854.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7b"}. This record does not assert independent reproduction.

### Search-o1

Average Score: 14.1. Result ID: 69bbb87247f34ce924a5a84f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76911e209131cea6dca9.md>)

[Implementation](<https://github.com/sunnynexus/Search-o1>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### ReAct

Average Score: 13.3. Result ID: 69bbb87247f34ce924a5a850.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### GSPO

Average Score: 11.7. Result ID: 69bbb87247f34ce924a5a846.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

### GRPO

Average Score: 11.3. Result ID: 69bbb87247f34ce924a5a845.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76a91e209131cea6e3eb.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

### Direct Inference

Average Score: 10.9. Result ID: 69bbb87247f34ce924a5a848.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-3b"}. This record does not assert independent reproduction.

### Search-o1

Average Score: 10. Result ID: 69bbb87247f34ce924a5a843.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76911e209131cea6dca9.md>)

[Implementation](<https://github.com/sunnynexus/Search-o1>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

### Direct Inference

Average Score: 8. Result ID: 69bbb87247f34ce924a5a84e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Llama3.2-3b"}. This record does not assert independent reproduction.

### ReAct

Average Score: 7.1. Result ID: 69bbb87247f34ce924a5a844.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

### Direct Inference

Average Score: 5.2. Result ID: 69bbb87247f34ce924a5a842.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c101e209131cea8c349.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-1.5b"}. This record does not assert independent reproduction.

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