# Question Answering on ARC Challenge (Action Rate and Risk Assessment)

Source: [Question Answering on ARC Challenge (Action Rate and Risk Assessment)](<https://www.sota2.com/research/sota/question-answering-on-arc-challenge-action-rate-and-risk-assessment>)

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

Data updated: 2026-05-21T10:27:21.108Z

Evaluation of model performance on the ARC Challenge dataset for Question Answering, specifically measuring Action Rate (AR), Risk, and Pathwise Violations (PathV).

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 (eval) |
| Metric | Pathwise Violations (PathV) |
| Unit | PERCENT |
| Direction | Lower 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

- [Pathwise Violations (PathV)](<https://www.sota2.com/research/sota/question-answering-on-arc-challenge-action-rate-and-risk-assessment.md?metric=Pathwise%20Violations%20(PathV)>): Lower is better; unit: PERCENT.

- [Action Rate (AR)](<https://www.sota2.com/research/sota/question-answering-on-arc-challenge-action-rate-and-risk-assessment.md?metric=Action%20Rate%20(AR)>): Higher is better; unit: PERCENT.

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

## Recorded results

| Method | Pathwise Violations (PathV) | Result date | Report paper |
| --- | --- | --- | --- |
| CSA | 0 | Not recorded | 6a0e918ff984739ed6738a88 |
| NEX-Conf | 20 | Not recorded | 6a0e918ff984739ed6738a88 |
| 5 online baselines | 100 | Not recorded | 6a0e918ff984739ed6738a88 |

### CSA

Pathwise Violations (PathV): 0. Result ID: 6a0eddf0fbc373592c06c3cc.

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

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

Recorded evaluation configuration: {"N (EVAL size)":"938","Err (Base verifier-fail rate)":"10.0%","alpha\* (Target risk level)":"0.10"}. This record does not assert independent reproduction.

### NEX-Conf

Pathwise Violations (PathV): 20. Result ID: 6a0eddf0fbc373592c06c3ce.

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

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

Recorded evaluation configuration: {"N (EVAL size)":"938","Err (Base verifier-fail rate)":"10.0%","alpha\* (Target risk level)":"0.10"}. This record does not assert independent reproduction.

### 5 online baselines

Pathwise Violations (PathV): 100. Result ID: 6a0eddf0fbc373592c06c3cd.

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

Recorded evaluation configuration: {"N (EVAL size)":"938","Err (Base verifier-fail rate)":"10.0%","alpha\* (Target risk level)":"0.10"}. This record does not assert independent reproduction.

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