# Question Answering on HotpotQA (EM and Latency)

Source: [Question Answering on HotpotQA (EM and Latency)](<https://www.sota2.com/research/sota/question-answering-on-hotpotqa-em-and-latency>)

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

Data updated: 2026-05-08T10:42:46.681Z

Evaluation of model performance on the HotpotQA dataset for Question Answering, measured using Exact Match (EM) and inference latency.

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 | Question Answering |
| Dataset (including any stated split) | HotpotQA |
| Metric | EM (%) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 27 of 27 |
| 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/question-answering-on-hotpotqa-em-and-latency.md?metric=EM%20(%25)>): Higher is better; unit: PERCENT.

- [Latency (ms)](<https://www.sota2.com/research/sota/question-answering-on-hotpotqa-em-and-latency.md?metric=Latency%20(ms)>): Lower is better; unit: TIME.

## Recorded results

| Method | EM (%) | Result date | Report paper |
| --- | --- | --- | --- |
| LatentRAG△ | 48.1 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 47.66 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 47.27 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 47.16 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 46.74 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 46.73 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 46.41 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 46.27 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 45.4 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 45.13 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 44.88 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 44.79 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 44.74 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 44.71 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 44.58 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 44.31 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 44.06 | Not recorded | 69fd6e205497b3b670c894ce |
| R1-Searcher | 43.34 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 43.05 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 42.63 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 41.99 | Not recorded | 69fd6e205497b3b670c894ce |
| DeepRAG | 33.44 | Not recorded | 69fd6e205497b3b670c894ce |
| Iter-RetGen | 31.74 | Not recorded | 69fd6e205497b3b670c894ce |
| ZeroSearch | 30.61 | Not recorded | 69fd6e205497b3b670c894ce |
| Naive RAG | 27.41 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-o1 | 25.09 | Not recorded | 69fd6e205497b3b670c894ce |
| Direct Infer | 16.68 | Not recorded | 69fd6e205497b3b670c894ce |

### LatentRAG△

EM (%): 48.1. Result ID: 69fdbe1d47f34ce924e16e62.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"jina-embeddings-v5-text-nano"}. This record does not assert independent reproduction.

### LatentRAG♢

EM (%): 47.66. Result ID: 69fdbe1d47f34ce924e16e60.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"jina-embeddings-v5-text-nano"}. This record does not assert independent reproduction.

### LatentRAG♢

EM (%): 47.27. Result ID: 69fdbe1d47f34ce924e16e64.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"harrier-oss-v1-270m"}. This record does not assert independent reproduction.

### LatentRAG△

EM (%): 47.16. Result ID: 69fdbe1d47f34ce924e16e5a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### LatentRAG△

EM (%): 46.74. Result ID: 69fdbe1d47f34ce924e16e66.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"harrier-oss-v1-270m"}. This record does not assert independent reproduction.

### LatentRAG♢

EM (%): 46.73. Result ID: 69fdbe1d47f34ce924e16e58.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### LatentRAG♢

EM (%): 46.41. Result ID: 69fdbe1d47f34ce924e16e5c.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"e5-base-v2"}. This record does not assert independent reproduction.

### LatentRAG△

EM (%): 46.27. Result ID: 69fdbe1d47f34ce924e16e5e.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"e5-base-v2"}. This record does not assert independent reproduction.

### AutoRefine△

EM (%): 45.4. Result ID: 69fdbe1d47f34ce924e16e65.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b641e209131cea88288.md>)

Recorded evaluation configuration: {"Retrieval Model":"harrier-oss-v1-270m"}. This record does not assert independent reproduction.

### AutoRefine△

EM (%): 45.13. Result ID: 69fdbe1d47f34ce924e16e5d.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b641e209131cea88288.md>)

Recorded evaluation configuration: {"Retrieval Model":"e5-base-v2"}. This record does not assert independent reproduction.

### Search-R1♢

EM (%): 44.88. Result ID: 69fdbe1d47f34ce924e16e63.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

Recorded evaluation configuration: {"Retrieval Model":"harrier-oss-v1-270m"}. This record does not assert independent reproduction.

### AutoRefine△

EM (%): 44.79. Result ID: 69fdbe1d47f34ce924e16e61.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b641e209131cea88288.md>)

Recorded evaluation configuration: {"Retrieval Model":"jina-embeddings-v5-text-nano"}. This record does not assert independent reproduction.

### LatentRAG△

EM (%): 44.74. Result ID: 69fdbe1d47f34ce924e16e6a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"F2LLM-v2-330M"}. This record does not assert independent reproduction.

### Search-R1♢

EM (%): 44.71. Result ID: 69fdbe1d47f34ce924e16e5b.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

Recorded evaluation configuration: {"Retrieval Model":"e5-base-v2"}. This record does not assert independent reproduction.

### LatentRAG♢

EM (%): 44.58. Result ID: 69fdbe1d47f34ce924e16e68.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"F2LLM-v2-330M"}. This record does not assert independent reproduction.

### Search-R1♢

EM (%): 44.31. Result ID: 69fdbe1d47f34ce924e16e5f.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

Recorded evaluation configuration: {"Retrieval Model":"jina-embeddings-v5-text-nano"}. This record does not assert independent reproduction.

### AutoRefine△

EM (%): 44.06. Result ID: 69fdbe1d47f34ce924e16e59.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b641e209131cea88288.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### R1-Searcher

EM (%): 43.34. Result ID: 69fdbe1d47f34ce924e16e54.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77771e209131cea7356d.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### Search-R1♢

EM (%): 43.05. Result ID: 69fdbe1d47f34ce924e16e57.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### AutoRefine△

EM (%): 42.63. Result ID: 69fdbe1d47f34ce924e16e69.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b641e209131cea88288.md>)

Recorded evaluation configuration: {"Retrieval Model":"F2LLM-v2-330M"}. This record does not assert independent reproduction.

### Search-R1♢

EM (%): 41.99. Result ID: 69fdbe1d47f34ce924e16e67.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

Recorded evaluation configuration: {"Retrieval Model":"F2LLM-v2-330M"}. This record does not assert independent reproduction.

### DeepRAG

EM (%): 33.44. Result ID: 69fdbe1d47f34ce924e16e56.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79411e209131cea7ce63.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### Iter-RetGen

EM (%): 31.74. Result ID: 69fdbe1d47f34ce924e16e52.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f21e209131cea4e48a.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### ZeroSearch

EM (%): 30.61. Result ID: 69fdbe1d47f34ce924e16e55.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa789d1e209131cea79568.md>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### Naive RAG

EM (%): 27.41. Result ID: 69fdbe1d47f34ce924e16e51.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b91e209131cea23c26.md>)

[Implementation](<https://github.com/huggingface/transformers/blob/master/examples/rag/>)

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### Search-o1

EM (%): 25.09. Result ID: 69fdbe1d47f34ce924e16e53.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

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

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

Recorded evaluation configuration: {"Retrieval Model":"Qwen3-Embedding-0.6B"}. This record does not assert independent reproduction.

### Direct Infer

EM (%): 16.68. Result ID: 69fdbe1d47f34ce924e16e50.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e205497b3b670c894ce.md>)

Recorded evaluation configuration: {"Retrieval Model":"N/A"}. This record does not assert independent reproduction.

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