# Long-context reasoning on OOLONG-REAL Average 650 samples

Source: [Long-context reasoning on OOLONG-REAL Average 650 samples](<https://www.sota2.com/research/sota/long-context-reasoning-on-oolong-real-average-650-samples>)

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

Data updated: 2026-05-08T11:42:23.813Z

Evaluation of model performance on long-context reasoning using the specific OOLONG-REAL subset, averaged over 650 samples, measured by Average reward.

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 | Long-context reasoning |
| Dataset (including any stated split) | OOLONG-REAL Average 650 samples |
| Metric | Average Reward |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Average Reward](<https://www.sota2.com/research/sota/long-context-reasoning-on-oolong-real-average-650-samples.md?metric=Average%20Reward>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Average Reward | Result date | Report paper |
| --- | --- | --- | --- |
| Recursive Agent Optimization | 0.32 | Not recorded | 69fd6e645497b3b670c897cd |
| Recursive Agent Optimization | 0.315 | Not recorded | 69fd6e645497b3b670c897cd |
| Single Agent | 0.203 | Not recorded | 69fd6e645497b3b670c897cd |
| Single Agent | 0.183 | Not recorded | 69fd6e645497b3b670c897cd |

### Recursive Agent Optimization

Average Reward: 0.32. Result ID: 69fdcc0e47f34ce924e1d056.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

Recorded evaluation configuration: {"Steps":"61.5","Time (s)":"175.4"}. This record does not assert independent reproduction.

### Recursive Agent Optimization

Average Reward: 0.315. Result ID: 69fdcc0e47f34ce924e1d05a.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

Recorded evaluation configuration: {"Steps":"61.5","Time (s)":"175.4"}. This record does not assert independent reproduction.

### Single Agent

Average Reward: 0.203. Result ID: 69fdcc0e47f34ce924e1d055.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

Recorded evaluation configuration: {"Steps":"7.1","Time (s)":"12.6"}. This record does not assert independent reproduction.

### Single Agent

Average Reward: 0.183. Result ID: 69fdcc0e47f34ce924e1d059.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6e645497b3b670c897cd.md>)

Recorded evaluation configuration: {"Steps":"7.1","Time (s)":"12.6"}. This record does not assert independent reproduction.

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