# Symbolic Reasoning on Symbolic Longer

Source: [Symbolic Reasoning on Symbolic Longer](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer>)

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

Data updated: 2026-02-26T05:34:13.448Z

Evaluation of model performance on the Symbolic Reasoning task using the 'Symbolic Longer' dataset, measured by detailed accuracy breakdowns across clean, irrelevant, and inaccurate conditions.

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 | Symbolic Reasoning |
| Dataset (including any stated split) | Symbolic Longer |
| Metric | Accuracy (Clean, Avg) |
| 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 (Clean, Avg)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Clean%2C%20Avg)>): Higher is better; unit: PERCENT.

- [Accuracy (Irrelevant, Easy)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Irrelevant%2C%20Easy)>): Higher is better; unit: PERCENT.

- [Accuracy (Irrelevant, Medium)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Irrelevant%2C%20Medium)>): Higher is better; unit: PERCENT.

- [Accuracy (Irrelevant, Hard)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Irrelevant%2C%20Hard)>): Higher is better; unit: PERCENT.

- [Accuracy (Irrelevant, Avg)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Irrelevant%2C%20Avg)>): Higher is better; unit: PERCENT.

- [Accuracy (Inaccurate, Easy)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Inaccurate%2C%20Easy)>): Higher is better; unit: PERCENT.

- [Accuracy (Inaccurate, Medium)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Inaccurate%2C%20Medium)>): Higher is better; unit: PERCENT.

- [Accuracy (Inaccurate, Hard)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Inaccurate%2C%20Hard)>): Higher is better; unit: PERCENT.

- [Accuracy (Inaccurate, Avg)](<https://www.sota2.com/research/sota/symbolic-reasoning-on-symbolic-longer.md?metric=Accuracy%20(Inaccurate%2C%20Avg)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (Clean, Avg) | Result date | Report paper |
| --- | --- | --- | --- |
| SCO | 0.187 | Not recorded | 68fa75741e209131cea6804d |
| CD-CoT | 0.123 | Not recorded | 68fa75741e209131cea6804d |
| CC | 0.094 | Not recorded | 68fa75741e209131cea6804d |
| Base | 0.092 | Not recorded | 68fa75741e209131cea6804d |
| BT | 0.072 | Not recorded | 68fa75741e209131cea6804d |

### SCO

Accuracy (Clean, Avg): 0.187. Result ID: 6993a6c33a6c314849178f09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72841e209131cea56f94.md>)

Recorded evaluation configuration: {"Additional Information":"Ground Truth"}. This record does not assert independent reproduction.

### CD-CoT

Accuracy (Clean, Avg): 0.123. Result ID: 6993a6c33a6c314849178f0c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

[Implementation](<https://github.com/tmlr-group/NoisyRationales>)

Recorded evaluation configuration: {"Additional Information":"Clean Demo"}. This record does not assert independent reproduction.

### CC

Accuracy (Clean, Avg): 0.094. Result ID: 6993a6c33a6c314849178f0b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa712b1e209131cea4fa78.md>)

Recorded evaluation configuration: {"Additional Information":"Clean Demo"}. This record does not assert independent reproduction.

### Base

Accuracy (Clean, Avg): 0.092. Result ID: 6993a6c33a6c314849178f08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

Recorded evaluation configuration: {"Additional Information":null}. This record does not assert independent reproduction.

### BT

Accuracy (Clean, Avg): 0.072. Result ID: 6993a6c33a6c314849178f0a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa75741e209131cea6804d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73771e209131cea5cb1a.md>)

[Implementation](<https://github.com/WHGTyen/BIG-Bench-Mistake>)

Recorded evaluation configuration: {"Additional Information":"Noise Position"}. This record does not assert independent reproduction.

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