# Natural Language Explanation Generation on ECQA few-shot 60-shot

Source: [Natural Language Explanation Generation on ECQA few-shot 60-shot](<https://www.sota2.com/research/sota/natural-language-explanation-generation-on-ecqa-few-shot-60-shot>)

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

Data updated: 2026-02-26T05:29:55.335Z

Evaluation of model performance on generating textual explanations using the 60-shot few-shot setting of the ECQA dataset, measured using Accuracy and nBERTs 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 | Natural Language Explanation Generation |
| Dataset (including any stated split) | ECQA few-shot 60-shot |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/natural-language-explanation-generation-on-ecqa-few-shot-60-shot.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [nBERTs](<https://www.sota2.com/research/sota/natural-language-explanation-generation-on-ecqa-few-shot-60-shot.md?metric=nBERTs>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SPARSEFIT | 24.53 | Not recorded | 68fa74701e209131cea61c52 |
| AdaLoRA | 13.04 | Not recorded | 68fa74701e209131cea61c52 |
| Full Fine-tuning | 11.14 | Not recorded | 68fa74701e209131cea61c52 |

### SPARSEFIT

Accuracy: 24.53. Result ID: 69935ccd3a6c3148491568db.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74701e209131cea61c52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74701e209131cea61c52.md>)

Recorded evaluation configuration: {"Fine-tuning Strategy":"Att.Q+LN","Backbone":"Llama 2-7B","Percentage Parameters":"7.97%","Few-shot setting":"60-shot"}. This record does not assert independent reproduction.

### AdaLoRA

Accuracy: 13.04. Result ID: 69935ccd3a6c3148491568dc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74701e209131cea61c52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71861e209131cea51c24.md>)

Recorded evaluation configuration: {"Fine-tuning Strategy":"PEFT","Backbone":"Llama 2-7B","Percentage Parameters":"0.30%","Few-shot setting":"60-shot"}. This record does not assert independent reproduction.

### Full Fine-tuning

Accuracy: 11.14. Result ID: 69935ccd3a6c3148491568da.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74701e209131cea61c52.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c661e209131cea33d9f.md>)

Recorded evaluation configuration: {"Fine-tuning Strategy":"Baseline","Backbone":"Llama 2-7B","Percentage Parameters":"100%","Few-shot setting":"60-shot"}. This record does not assert independent reproduction.

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