# Logical Reasoning on RP Npred ≤ 30, 6 \< δ ≤ 12 (train)

Source: [Logical Reasoning on RP Npred ≤ 30, 6 \< δ ≤ 12 (train)](<https://www.sota2.com/research/sota/logical-reasoning-on-rp-npred-30-6-12-train>)

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

Data updated: 2026-05-07T06:55:46.503Z

Evaluation of model performance on Logical Reasoning using the training subset of the RP dataset defined by Npred ≤ 30 and 6 \< δ ≤ 12, 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 | Logical Reasoning |
| Dataset (including any stated split) | RP Npred ≤ 30, 6 \< δ ≤ 12 (train) |
| Metric | Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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 (RP Npred ≤ 30, 6 \< δ ≤ 12 train)](<https://www.sota2.com/research/sota/logical-reasoning-on-rp-npred-30-6-12-train.md?metric=Accuracy%20(RP%20Npred%20%E2%89%A4%2030%2C%206%20%3C%20%CE%B4%20%E2%89%A4%2012%20train)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train) | Result date | Report paper |
| --- | --- | --- | --- |
| cot w. L = 64 | 99.5 | Not recorded | 69fc1c725497b3b670c88433 |
| cot w. L = 128 | 99.4 | Not recorded | 69fc1c725497b3b670c88433 |
| direct w. L = 128 | 98.5 | Not recorded | 69fc1c725497b3b670c88433 |
| cot (baseline) | 96.8 | Not recorded | 69fc1c725497b3b670c88433 |
| direct (baseline) | 72.9 | Not recorded | 69fc1c725497b3b670c88433 |

### cot w. L = 64

Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train): 99.5. Result ID: 69fc368947f34ce924de835e.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

Recorded evaluation configuration: {"Reasoning Strategy":"CoT","Number of Layers":"64"}. This record does not assert independent reproduction.

### cot w. L = 128

Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train): 99.4. Result ID: 69fc368947f34ce924de835f.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

Recorded evaluation configuration: {"Reasoning Strategy":"CoT","Number of Layers":"128"}. This record does not assert independent reproduction.

### direct w. L = 128

Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train): 98.5. Result ID: 69fc368947f34ce924de835c.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

Recorded evaluation configuration: {"Reasoning Strategy":"Direct","Number of Layers":"128"}. This record does not assert independent reproduction.

### cot (baseline)

Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train): 96.8. Result ID: 69fc368947f34ce924de835d.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

Recorded evaluation configuration: {"Reasoning Strategy":"CoT"}. This record does not assert independent reproduction.

### direct (baseline)

Accuracy (RP Npred ≤ 30, 6 \< δ ≤ 12 train): 72.9. Result ID: 69fc368947f34ce924de835b.

[Reporting paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fc1c725497b3b670c88433.md>)

Recorded evaluation configuration: {"Reasoning Strategy":"Direct"}. This record does not assert independent reproduction.

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