# Extractive Question Answering on TriviaQA 20% coverage

Source: [Extractive Question Answering on TriviaQA 20% coverage](<https://www.sota2.com/research/sota/extractive-question-answering-on-triviaqa-20-percent-coverage>)

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

Data updated: 2026-02-26T05:42:45.618Z

Evaluation of model performance on the 20% coverage subset of the TriviaQA dataset for Extractive Question Answering, measured by F1 score.

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 | Extractive Question Answering |
| Dataset (including any stated split) | TriviaQA 20% coverage |
| Metric | F1 |
| 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

- [F1](<https://www.sota2.com/research/sota/extractive-question-answering-on-triviaqa-20-percent-coverage.md?metric=F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 | Result date | Report paper |
| --- | --- | --- | --- |
| QA+E+C | 81.33 | Not recorded | 68fa6f8e1e209131cea463a9 |
| QA+E | 81.21 | Not recorded | 68fa6f8e1e209131cea463a9 |
| QA | 80.68 | Not recorded | 68fa6f8e1e209131cea463a9 |
| QA+C | 80.36 | Not recorded | 68fa6f8e1e209131cea463a9 |
| C | 69.13 | Not recorded | 68fa6f8e1e209131cea463a9 |
| E+C | 65.53 | Not recorded | 68fa6f8e1e209131cea463a9 |
| E | 65.25 | Not recorded | 68fa6f8e1e209131cea463a9 |

### QA+E+C

F1: 81.33. Result ID: 6990dee13a6c31484903eb37.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

Recorded evaluation configuration: {"Selection Strategy":"QA+E+C","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### QA+E

F1: 81.21. Result ID: 6990dee13a6c31484903eb39.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

Recorded evaluation configuration: {"Selection Strategy":"QA+E","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### QA

F1: 80.68. Result ID: 6990dee13a6c31484903eb3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

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

Recorded evaluation configuration: {"Selection Strategy":"QA","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### QA+C

F1: 80.36. Result ID: 6990dee13a6c31484903eb38.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

Recorded evaluation configuration: {"Selection Strategy":"QA+C","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### C

F1: 69.13. Result ID: 6990dee13a6c31484903eb3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aef1e209131cea2b2a6.md>)

[Implementation](<https://github.com/microsoft/DeBERTa111>)

Recorded evaluation configuration: {"Selection Strategy":"C","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### E+C

F1: 65.53. Result ID: 6990dee13a6c31484903eb3a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aef1e209131cea2b2a6.md>)

[Implementation](<https://github.com/microsoft/DeBERTa111>)

Recorded evaluation configuration: {"Selection Strategy":"E+C","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

### E

F1: 65.25. Result ID: 6990dee13a6c31484903eb3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f8e1e209131cea463a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6aef1e209131cea2b2a6.md>)

[Implementation](<https://github.com/microsoft/DeBERTa111>)

Recorded evaluation configuration: {"Selection Strategy":"E","Coverage":"20%","Backbone":"BERT-large","Domain":"Out-of-domain"}. This record does not assert independent reproduction.

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