# Question Answering on PopQA (EM(%) and Latency)

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

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

Data updated: 2026-05-08T10:42:46.637Z

Evaluation of model performance on the PopQA dataset for Question Answering, specifically reporting Exact Match accuracy 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) | PopQA |
| 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-popqa-em-percent-and-latency.md?metric=EM%20(%25)>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | EM (%) | Result date | Report paper |
| --- | --- | --- | --- |
| LatentRAG△ | 47.82 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 47.68 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 47.42 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 47.31 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 47.19 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 47.14 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 47.12 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 46.84 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 46.72 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 46.63 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 46.54 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 46.02 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 45.92 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 45.41 | Not recorded | 69fd6e205497b3b670c894ce |
| AutoRefine△ | 45.17 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-R1♢ | 44.92 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 44.72 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 44.21 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG△ | 43.88 | Not recorded | 69fd6e205497b3b670c894ce |
| LatentRAG♢ | 43.04 | Not recorded | 69fd6e205497b3b670c894ce |
| ZeroSearch | 40.72 | Not recorded | 69fd6e205497b3b670c894ce |
| Iter-RetGen | 40.29 | Not recorded | 69fd6e205497b3b670c894ce |
| DeepRAG | 39.15 | Not recorded | 69fd6e205497b3b670c894ce |
| R1-Searcher | 38.37 | Not recorded | 69fd6e205497b3b670c894ce |
| Naive RAG | 35.72 | Not recorded | 69fd6e205497b3b670c894ce |
| Search-o1 | 31.25 | Not recorded | 69fd6e205497b3b670c894ce |
| Direct Infer | 14.5 | Not recorded | 69fd6e205497b3b670c894ce |

### LatentRAG△

EM (%): 47.82. Result ID: 69fdbe1d47f34ce924e16e47.

[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.

### AutoRefine△

EM (%): 47.68. Result ID: 69fdbe1d47f34ce924e16e4a.

[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.

### LatentRAG△

EM (%): 47.42. Result ID: 69fdbe1d47f34ce924e16e3f.

[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.

### Search-R1♢

EM (%): 47.31. Result ID: 69fdbe1d47f34ce924e16e3c.

[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 (%): 47.19. Result ID: 69fdbe1d47f34ce924e16e3e.

[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.

### Search-R1♢

EM (%): 47.14. Result ID: 69fdbe1d47f34ce924e16e48.

[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 (%): 47.12. Result ID: 69fdbe1d47f34ce924e16e42.

[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.

### AutoRefine△

EM (%): 46.84. Result ID: 69fdbe1d47f34ce924e16e46.

[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 (%): 46.72. Result ID: 69fdbe1d47f34ce924e16e45.

[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 (%): 46.63. Result ID: 69fdbe1d47f34ce924e16e3d.

[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.

### Search-R1♢

EM (%): 46.54. Result ID: 69fdbe1d47f34ce924e16e44.

[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.

### Search-R1♢

EM (%): 46.02. Result ID: 69fdbe1d47f34ce924e16e40.

[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 (%): 45.92. Result ID: 69fdbe1d47f34ce924e16e4b.

[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 (%): 45.41. Result ID: 69fdbe1d47f34ce924e16e49.

[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.

### AutoRefine△

EM (%): 45.17. Result ID: 69fdbe1d47f34ce924e16e4e.

[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 (%): 44.92. Result ID: 69fdbe1d47f34ce924e16e4c.

[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.

### LatentRAG△

EM (%): 44.72. Result ID: 69fdbe1d47f34ce924e16e4f.

[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.

### LatentRAG♢

EM (%): 44.21. Result ID: 69fdbe1d47f34ce924e16e4d.

[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.

### LatentRAG△

EM (%): 43.88. Result ID: 69fdbe1d47f34ce924e16e43.

[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 (%): 43.04. Result ID: 69fdbe1d47f34ce924e16e41.

[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.

### ZeroSearch

EM (%): 40.72. Result ID: 69fdbe1d47f34ce924e16e3a.

[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.

### Iter-RetGen

EM (%): 40.29. Result ID: 69fdbe1d47f34ce924e16e37.

[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.

### DeepRAG

EM (%): 39.15. Result ID: 69fdbe1d47f34ce924e16e3b.

[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.

### R1-Searcher

EM (%): 38.37. Result ID: 69fdbe1d47f34ce924e16e39.

[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.

### Naive RAG

EM (%): 35.72. Result ID: 69fdbe1d47f34ce924e16e36.

[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 (%): 31.25. Result ID: 69fdbe1d47f34ce924e16e38.

[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 (%): 14.5. Result ID: 69fdbe1d47f34ce924e16e35.

[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.
