# Temporal Question Answering on ReasonQA Single-hop

Source: [Temporal Question Answering on ReasonQA Single-hop](<https://www.sota2.com/research/sota/temporal-question-answering-on-reasonqa-single-hop>)

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

Data updated: 2026-02-26T06:04:18.726Z

Evaluation of model performance on the ReasonQA Single-hop dataset for Temporal Question Answering, reporting Set Accuracy and Answer F1 scores.

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 | Temporal Question Answering |
| Dataset (including any stated split) | ReasonQA Single-hop |
| Metric | Set Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Set Accuracy](<https://www.sota2.com/research/sota/temporal-question-answering-on-reasonqa-single-hop.md?metric=Set%20Accuracy>): Higher is better; unit: PERCENT.

- [Answer F1](<https://www.sota2.com/research/sota/temporal-question-answering-on-reasonqa-single-hop.md?metric=Answer%20F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Set Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| T5-large PIT-SFT | 95.1 | Not recorded | 68fa740c1e209131cea5f719 |
| T5-base PIT-SFT | 91.6 | Not recorded | 68fa740c1e209131cea5f719 |
| T5-large SFT | 86 | Not recorded | 68fa740c1e209131cea5f719 |
| T5-base SFT | 80.4 | Not recorded | 68fa740c1e209131cea5f719 |
| GPT-4 | 67.1 | Not recorded | 68fa740c1e209131cea5f719 |
| FLAN-T5-XL | 61.5 | Not recorded | 68fa740c1e209131cea5f719 |
| GPT-3.5 | 28 | Not recorded | 68fa740c1e209131cea5f719 |

### T5-large PIT-SFT

Set Accuracy: 95.1. Result ID: 699316253a6c314849139c98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Implementation](<https://github.com/nusnlp/complex-tr>)

Recorded evaluation configuration: {"Backbone":"T5-large","Training strategy":"PIT-SFT"}. This record does not assert independent reproduction.

### T5-base PIT-SFT

Set Accuracy: 91.6. Result ID: 699316253a6c314849139c96.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Implementation](<https://github.com/nusnlp/complex-tr>)

Recorded evaluation configuration: {"Backbone":"T5-base","Training strategy":"PIT-SFT"}. This record does not assert independent reproduction.

### T5-large SFT

Set Accuracy: 86. Result ID: 699316253a6c314849139c97.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa709e1e209131cea4c470.md>)

[Implementation](<https://github.com/google-research/text-to-text-transfer-transformer>)

Recorded evaluation configuration: {"Backbone":"T5-large","Training strategy":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### T5-base SFT

Set Accuracy: 80.4. Result ID: 699316253a6c314849139c95.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa709e1e209131cea4c470.md>)

[Implementation](<https://github.com/google-research/text-to-text-transfer-transformer>)

Recorded evaluation configuration: {"Backbone":"T5-base","Training strategy":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### GPT-4

Set Accuracy: 67.1. Result ID: 699316253a6c314849139c94.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

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

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

Recorded evaluation configuration: {"Model version":"gpt4-0613","Few-shot":"one-shot"}. This record does not assert independent reproduction.

### FLAN-T5-XL

Set Accuracy: 61.5. Result ID: 699316253a6c314849139c92.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bc51e209131cea30383.md>)

[Implementation](<https://github.com/google-research/flan>)

Recorded evaluation configuration: {"Parameters":"3B","Few-shot":"one-shot"}. This record does not assert independent reproduction.

### GPT-3.5

Set Accuracy: 28. Result ID: 699316253a6c314849139c93.

[Reporting paper](<https://www.sota2.com/research/paper/68fa740c1e209131cea5f719.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bf71e209131cea314c0.md>)

[Implementation](<https://github.com/openai/following-instructions-human-feedback>)

Recorded evaluation configuration: {"Model version":"gpt3.5-turbo","Few-shot":"one-shot"}. This record does not assert independent reproduction.

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