# Multi-hop Question Answering on HOTPOTQA (EM, P, R, F1, BSc, HSc)

Source: [Multi-hop Question Answering on HOTPOTQA (EM, P, R, F1, BSc, HSc)](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc>)

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

Data updated: 2026-04-14T08:30:53.341Z

Evaluation of model performance on the HotpotQA dataset for multi-hop question answering, reporting Exact Match (EM), Precision (P), Recall (R), F1 score, Boundary Score (BSc), and Hypothesis Score (HSc).

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) | HOTPOTQA |
| Metric | EM |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=EM>): Higher is better; unit: PERCENT.

- [P](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=P>): Higher is better; unit: PERCENT.

- [R](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=R>): Higher is better; unit: PERCENT.

- [F1](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=F1>): Higher is better; unit: PERCENT.

- [BSc](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=BSc>): Higher is better; unit: PERCENT.

- [HSc](<https://www.sota2.com/research/sota/multi-hop-question-answering-on-hotpotqa-em-p-r-f1-bsc-hsc.md?metric=HSc>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EM | Result date | Report paper |
| --- | --- | --- | --- |
| KCS | 61 | Not recorded | 68fa7b001e209131cea862a6 |
| RAST | 55.5 | Not recorded | 68fa7b001e209131cea862a6 |
| Composition Sampling | 55 | Not recorded | 68fa7b001e209131cea862a6 |
| Original | 54 | Not recorded | 68fa7b001e209131cea862a6 |
| GPT4 | 49.5 | Not recorded | 68fa7b001e209131cea862a6 |
| KCS | 45 | Not recorded | 68fa7b001e209131cea862a6 |
| Composition Sampling | 40.5 | Not recorded | 68fa7b001e209131cea862a6 |
| Original | 39.5 | Not recorded | 68fa7b001e209131cea862a6 |
| RAST | 34.5 | Not recorded | 68fa7b001e209131cea862a6 |
| GPT-4 | 32.5 | Not recorded | 68fa7b001e209131cea862a6 |

### KCS

EM: 61. Result ID: 69ddf9d547f34ce924c55d51.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

Recorded evaluation configuration: {"LLM":"LLAMA3.1","Diversifying Method":"+KCS"}. This record does not assert independent reproduction.

### RAST

EM: 55.5. Result ID: 69ddf9d547f34ce924c55d4e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f41e209131cea4e61a.md>)

Recorded evaluation configuration: {"LLM":"LLAMA3.1","Diversifying Method":"+RAST"}. This record does not assert independent reproduction.

### Composition Sampling

EM: 55. Result ID: 69ddf9d547f34ce924c55d4f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c161e209131cea320bd.md>)

[Implementation](<https://github.com/google-research/language/tree/master/language/frost>)

Recorded evaluation configuration: {"LLM":"LLAMA3.1","Diversifying Method":"+Composition"}. This record does not assert independent reproduction.

### Original

EM: 54. Result ID: 69ddf9d547f34ce924c55d4d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

Recorded evaluation configuration: {"LLM":"LLAMA3.1","Diversifying Method":"ORI"}. This record does not assert independent reproduction.

### GPT4

EM: 49.5. Result ID: 69ddf9d547f34ce924c55d50.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"LLM":"LLAMA3.1","Diversifying Method":"+GPT4"}. This record does not assert independent reproduction.

### KCS

EM: 45. Result ID: 69ddf9d547f34ce924c55d56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

Recorded evaluation configuration: {"LLM":"QWEN2.5","Diversifying Method":"+KCS"}. This record does not assert independent reproduction.

### Composition Sampling

EM: 40.5. Result ID: 69ddf9d547f34ce924c55d54.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c161e209131cea320bd.md>)

[Implementation](<https://github.com/google-research/language/tree/master/language/frost>)

Recorded evaluation configuration: {"LLM":"QWEN2.5","Diversifying Method":"+Composition"}. This record does not assert independent reproduction.

### Original

EM: 39.5. Result ID: 69ddf9d547f34ce924c55d52.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

Recorded evaluation configuration: {"LLM":"QWEN2.5","Diversifying Method":"ORI"}. This record does not assert independent reproduction.

### RAST

EM: 34.5. Result ID: 69ddf9d547f34ce924c55d53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f41e209131cea4e61a.md>)

Recorded evaluation configuration: {"LLM":"QWEN2.5","Diversifying Method":"+RAST"}. This record does not assert independent reproduction.

### GPT-4

EM: 32.5. Result ID: 69ddf9d547f34ce924c55d55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b001e209131cea862a6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"LLM":"QWEN2.5","Diversifying Method":"+GPT-4"}. This record does not assert independent reproduction.

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