# Single-hop QA on SimpleQuestions

Source: [Single-hop QA on SimpleQuestions](<https://www.sota2.com/research/sota/single-hop-qa-on-simplequestions>)

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

Data updated: 2026-02-26T05:31:10.330Z

Evaluation of model performance on the SimpleQuestions dataset for single-hop question answering tasks, reporting accuracy.

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 | Single-hop QA |
| Dataset (including any stated split) | SimpleQuestions |
| Metric | Accuracy |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/single-hop-qa-on-simplequestions.md?metric=Accuracy>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4+RFKG-CoT | 0.87 | Not recorded | 69438a74423ea422657293a2 |
| GPT-4+KG-CoT | 0.861 | Not recorded | 69438a74423ea422657293a2 |
| DiFaR | 0.858 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+RFKG-CoT | 0.792 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+KG-CoT | 0.778 | Not recorded | 69438a74423ea422657293a2 |
| RNG | 0.711 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+CoT | 0.203 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+IO | 0.2 | Not recorded | 69438a74423ea422657293a2 |
| ChatGPT+SC | 0.189 | Not recorded | 69438a74423ea422657293a2 |

### GPT-4+RFKG-CoT

Accuracy: 0.87. Result ID: 6996bced5698665ec407cf8b.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"GPT-4"}. This record does not assert independent reproduction.

### GPT-4+KG-CoT

Accuracy: 0.861. Result ID: 6996bced5698665ec407cf89.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"GPT-4"}. This record does not assert independent reproduction.

### DiFaR

Accuracy: 0.858. Result ID: 6996bced5698665ec407cf86.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f641e209131cea45413.md>)

Recorded evaluation configuration: {"AccessKB":true}. This record does not assert independent reproduction.

### ChatGPT+RFKG-CoT

Accuracy: 0.792. Result ID: 6996bced5698665ec407cf8a.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+KG-CoT

Accuracy: 0.778. Result ID: 6996bced5698665ec407cf88.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

Recorded evaluation configuration: {"AccessKB":true,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### RNG

Accuracy: 0.711. Result ID: 6996bced5698665ec407cf87.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c131e209131cea31f4b.md>)

Recorded evaluation configuration: {"AccessKB":true}. This record does not assert independent reproduction.

### ChatGPT+CoT

Accuracy: 0.203. Result ID: 6996bced5698665ec407cf84.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e571e209131cea3f27b.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+IO

Accuracy: 0.2. Result ID: 6996bced5698665ec407cf83.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7e1e209131cea404d4.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

### ChatGPT+SC

Accuracy: 0.189. Result ID: 6996bced5698665ec407cf85.

[Reporting paper](<https://www.sota2.com/research/paper/69438a74423ea422657293a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb61e209131cea41b33.md>)

Recorded evaluation configuration: {"AccessKB":false,"Backbone":"ChatGPT"}. This record does not assert independent reproduction.

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