# Question Answering on ARC Easy DE 25-shot (test)

Source: [Question Answering on ARC Easy DE 25-shot (test)](<https://www.sota2.com/research/sota/question-answering-on-arc-easy-de-25-shot-test>)

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

Data updated: 2026-03-18T07:37:15.153Z

Evaluation of model performance on the 25-shot test split of the ARC Easy DE dataset for Question Answering, measured by normalized log accuracy and sequence efficiency.

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 | Question Answering |
| Dataset (including any stated split) | ARC Easy DE 25-shot (test) |
| Metric | Normalized Log 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

- [Normalized Log Accuracy](<https://www.sota2.com/research/sota/question-answering-on-arc-easy-de-25-shot-test.md?metric=Normalized%20Log%20Accuracy>): Higher is better; unit: PERCENT.

- [Bytes per Sequence Position](<https://www.sota2.com/research/sota/question-answering-on-arc-easy-de-25-shot-test.md?metric=Bytes%20per%20Sequence%20Position>): Lower is better; unit: SIZE.

## Recorded results

| Method | Normalized Log Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| T-Free | 80.1 | Not recorded | 69ba317b0636e5bf4afc071d |
| HATified | 77.9 | Not recorded | 69ba317b0636e5bf4afc071d |
| Llama | 71.5 | Not recorded | 69ba317b0636e5bf4afc071d |

### T-Free

Normalized Log Accuracy: 80.1. Result ID: 69ba561247f34ce924a3d954.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc071d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc071d.md>)

Recorded evaluation configuration: {"shots":"25-shot","evaluation\_mode":"Pre-training"}. This record does not assert independent reproduction.

### HATified

Normalized Log Accuracy: 77.9. Result ID: 69ba561247f34ce924a3d955.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc071d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc071d.md>)

Recorded evaluation configuration: {"shots":"25-shot","evaluation\_mode":"Pre-training"}. This record does not assert independent reproduction.

### Llama

Normalized Log Accuracy: 71.5. Result ID: 69ba561247f34ce924a3d956.

[Reporting paper](<https://www.sota2.com/research/paper/69ba317b0636e5bf4afc071d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"shots":"25-shot","evaluation\_mode":"Pre-training"}. This record does not assert independent reproduction.

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