# Reasoning Question Answering on ARC Challenge

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

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

Data updated: 2026-05-07T09:26:36.667Z

Evaluation of model performance on the ARC Challenge dataset focusing specifically on reasoning-based question answering, typically measured by 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 | Reasoning Question Answering |
| Dataset (including any stated split) | ARC Challenge |
| Metric | 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

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

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| LoPT-GRPO | 74.55 | Not recorded | 69fc1c8e5497b3b670c88507 |
| E2E-GRPO | 74.23 | Not recorded | 69fc1c8e5497b3b670c88507 |
| Base | 74.06 | Not recorded | 69fc1c8e5497b3b670c88507 |

### LoPT-GRPO

Accuracy: 74.55. Result ID: 69fc5a7747f34ce924df5b3f.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c8e5497b3b670c88507.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c8e5497b3b670c88507.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-32B"}. This record does not assert independent reproduction.

### E2E-GRPO

Accuracy: 74.23. Result ID: 69fc5a7747f34ce924df5b3e.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c8e5497b3b670c88507.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-32B"}. This record does not assert independent reproduction.

### Base

Accuracy: 74.06. Result ID: 69fc5a7747f34ce924df5b3d.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c8e5497b3b670c88507.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-32B"}. This record does not assert independent reproduction.

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