# Generative Question Answering on SQuAD Del-Word (test)

Source: [Generative Question Answering on SQuAD Del-Word (test)](<https://www.sota2.com/research/sota/generative-question-answering-on-squad-del-word-test>)

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

Data updated: 2026-06-11T08:05:30.318Z

Evaluation of model performance on the test split of the SQuAD Del-Word dataset using Generative Question Answering metrics, reporting EM and 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 | Generative Question Answering |
| Dataset (including any stated split) | SQuAD Del-Word (test) |
| Metric | EM |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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/generative-question-answering-on-squad-del-word-test.md?metric=EM>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/generative-question-answering-on-squad-del-word-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | EM | Result date | Report paper |
| --- | --- | --- | --- |
| SDBN | 51.7 | Not recorded | 6a28f37090966ade919c1925 |
| Vanilla | 49.3 | Not recorded | 6a28f37090966ade919c1925 |
| NEFTune | 48.9 | Not recorded | 6a28f37090966ade919c1925 |

### SDBN

EM: 51.7. Result ID: 6a2a6c11fbc373592c275df1.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f37090966ade919c1925.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a28f37090966ade919c1925.md>)

Recorded evaluation configuration: {"Backbone":"DeBERTa-v3","Training Samples":"500","Average Runs":"5"}. This record does not assert independent reproduction.

### Vanilla

EM: 49.3. Result ID: 6a2a6c11fbc373592c275df0.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f37090966ade919c1925.md>)

Recorded evaluation configuration: {"Backbone":"DeBERTa-v3","Training Samples":"500","Average Runs":"5"}. This record does not assert independent reproduction.

### NEFTune

EM: 48.9. Result ID: 6a2a6c11fbc373592c275df2.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f37090966ade919c1925.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cd1e209131cea4d3fc.md>)

Recorded evaluation configuration: {"Backbone":"DeBERTa-v3","Training Samples":"500","Average Runs":"5"}. This record does not assert independent reproduction.

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