# Question Answering on ARC (acc\_norm)

Source: [Question Answering on ARC (acc\_norm)](<https://www.sota2.com/research/sota/question-answering-on-arc-acc_norm>)

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

Data updated: 2026-04-23T10:58:16.095Z

Evaluation of model performance on the Abstract Reasoning Corpus (ARC) for Question Answering, measured using normalized accuracy (acc\_norm).

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 |
| Metric | Normalized 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 Accuracy](<https://www.sota2.com/research/sota/question-answering-on-arc-acc_norm.md?metric=Normalized%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Normalized Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| 1B Model (Multilingual classifier) | 31.71 | Not recorded | 69e9a78a55a3c32af89f7037 |
| 1B Model (HQ filtering) | 31.45 | Not recorded | 69e9a78a55a3c32af89f7037 |
| 1B Model (No filtering) | 28.55 | Not recorded | 69e9a78a55a3c32af89f7037 |

### 1B Model (Multilingual classifier)

Normalized Accuracy: 31.71. Result ID: 69e9fa8247f34ce924d1fbb1.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78a55a3c32af89f7037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e9a78a55a3c32af89f7037.md>)

Recorded evaluation configuration: {"Filtering Strategy":"Multilingual (ML)","Model Scale":"1B"}. This record does not assert independent reproduction.

### 1B Model (HQ filtering)

Normalized Accuracy: 31.45. Result ID: 69e9fa8247f34ce924d1fbb0.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78a55a3c32af89f7037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b6c1e209131cea885c7.md>)

Recorded evaluation configuration: {"Filtering Strategy":"Monolingual high-quality (HQ)","Model Scale":"1B"}. This record does not assert independent reproduction.

### 1B Model (No filtering)

Normalized Accuracy: 28.55. Result ID: 69e9fa8247f34ce924d1fbaf.

[Reporting paper](<https://www.sota2.com/research/paper/69e9a78a55a3c32af89f7037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b6c1e209131cea885c7.md>)

Recorded evaluation configuration: {"Filtering Strategy":"No filtering","Model Scale":"1B"}. This record does not assert independent reproduction.

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