# Question Answering on ARC (AR% and Risk%)

Source: [Question Answering on ARC (AR% and Risk%)](<https://www.sota2.com/research/sota/question-answering-on-arc-ar-percent-and-risk-percent>)

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

Data updated: 2026-05-21T10:27:20.891Z

Evaluation of model performance on the Abstract Reasoning Corpus (ARC) for Question Answering, specifically tracking AR% and Risk% metrics.

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 | AR% |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [AR%](<https://www.sota2.com/research/sota/question-answering-on-arc-ar-percent-and-risk-percent.md?metric=AR%25>): Higher is better; unit: PERCENT.

- [Risk%](<https://www.sota2.com/research/sota/question-answering-on-arc-ar-percent-and-risk-percent.md?metric=Risk%25>): Lower is better; unit: PERCENT.

## Recorded results

| Method | AR% | Result date | Report paper |
| --- | --- | --- | --- |
| Fixed-Threshold | 100 | Not recorded | 6a0e918ff984739ed6738a88 |
| Naive-Tuning | 100 | Not recorded | 6a0e918ff984739ed6738a88 |
| Always-Act | 100 | Not recorded | 6a0e918ff984739ed6738a88 |
| ACI | 99.9 | Not recorded | 6a0e918ff984739ed6738a88 |
| SAOCP | 99.3 | Not recorded | 6a0e918ff984739ed6738a88 |
| NEX-Conf | 97.8 | Not recorded | 6a0e918ff984739ed6738a88 |
| CSA | 56.2 | Not recorded | 6a0e918ff984739ed6738a88 |

### Fixed-Threshold

AR%: 100. Result ID: 6a0eddf0fbc373592c06c3a3.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### Naive-Tuning

AR%: 100. Result ID: 6a0eddf0fbc373592c06c3a4.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### Always-Act

AR%: 100. Result ID: 6a0eddf0fbc373592c06c3a5.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### ACI

AR%: 99.9. Result ID: 6a0eddf0fbc373592c06c3a2.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b601e209131cea2dd5d.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### SAOCP

AR%: 99.3. Result ID: 6a0eddf0fbc373592c06c3a1.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5a0f1e209131ce9bc942.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### NEX-Conf

AR%: 97.8. Result ID: 6a0eddf0fbc373592c06c3a0.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7e861e209131cea96bce.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

### CSA

AR%: 56.2. Result ID: 6a0eddf0fbc373592c06c39f.

[Reporting paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0e918ff984739ed6738a88.md>)

Recorded evaluation configuration: {"alpha":"0.10"}. This record does not assert independent reproduction.

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