# Question Answering on ARC Challenge DE

Source: [Question Answering on ARC Challenge DE](<https://www.sota2.com/research/sota/question-answering-on-arc-challenge-de>)

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

Data updated: 2026-03-18T07:37:14.575Z

Evaluation of model performance on the ARC Challenge DE dataset for Question Answering, measured using Normalized Log Accuracy.

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 Challenge DE |
| Metric | Normalized Log Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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-challenge-de.md?metric=Normalized%20Log%20Accuracy>): Higher is better; unit: PERCENT.

- [Compression](<https://www.sota2.com/research/sota/question-answering-on-arc-challenge-de.md?metric=Compression>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Normalized Log Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| HATified-SFT | 66.9 | Not recorded | 69ba317b0636e5bf4afc071d |
| Llama-Instruct | 65.3 | Not recorded | 69ba317b0636e5bf4afc071d |
| T-Free | 64 | Not recorded | 69ba317b0636e5bf4afc071d |
| HATified | 59.4 | Not recorded | 69ba317b0636e5bf4afc071d |
| Llama | 51.5 | Not recorded | 69ba317b0636e5bf4afc071d |
| Tülu | 36.3 | Not recorded | 69ba317b0636e5bf4afc071d |

### HATified-SFT

Normalized Log Accuracy: 66.9. Result ID: 69ba561247f34ce924a3d8e5.

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

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

Recorded evaluation configuration: {"Mode":"SFT","Shots":"25-shot"}. This record does not assert independent reproduction.

### Llama-Instruct

Normalized Log Accuracy: 65.3. Result ID: 69ba561247f34ce924a3d8e6.

[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: {"Mode":"SFT","Shots":"25-shot"}. This record does not assert independent reproduction.

### T-Free

Normalized Log Accuracy: 64. Result ID: 69ba561247f34ce924a3d8ff.

[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"}. This record does not assert independent reproduction.

### HATified

Normalized Log Accuracy: 59.4. Result ID: 69ba561247f34ce924a3d900.

[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"}. This record does not assert independent reproduction.

### Llama

Normalized Log Accuracy: 51.5. Result ID: 69ba561247f34ce924a3d901.

[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"}. This record does not assert independent reproduction.

### Tülu

Normalized Log Accuracy: 36.3. Result ID: 69ba561247f34ce924a3d902.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa78121e209131cea76a04.md>)

Recorded evaluation configuration: {"shots":"25-shot"}. This record does not assert independent reproduction.

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