# Long-horizon forecasting on KS (test)

Source: [Long-horizon forecasting on KS (test)](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test>)

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

Data updated: 2026-02-26T06:11:26.515Z

Evaluation of Long-horizon forecasting performance on the test split of the KS dataset, measured using RMSE, ACC, and Specific Error metrics.

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-horizon forecasting |
| Dataset (including any stated split) | KS (test) |
| Metric | ∆RMSE (%) |
| 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

- [∆RMSE (%)](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=%E2%88%86RMSE%20(%25)>): Lower is better; unit: PERCENT.

- [RMSE @ H](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=RMSE%20%40%20H>): Lower is better; unit: SCALAR.

- [ACC @ H](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=ACC%20%40%20H>): Higher is better; unit: PERCENT.

- [Specific Error](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=Specific%20Error>): Lower is better; unit: SCALAR.

- [∆ACC (%)](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=%E2%88%86ACC%20(%25)>): Higher is better; unit: PERCENT.

- [∆SpecErr (%)](<https://www.sota2.com/research/sota/long-horizon-forecasting-on-ks-test.md?metric=%E2%88%86SpecErr%20(%25)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | ∆RMSE (%) | Result date | Report paper |
| --- | --- | --- | --- |
| U-Net-AR | 0 | Not recorded | 6971aefee25ad2fa9d1ed8d1 |
| HINE-L2 | 34.7 | Not recorded | 6971aefee25ad2fa9d1ed8d1 |
| MSR-HINE | 62.8 | Not recorded | 6971aefee25ad2fa9d1ed8d1 |

### U-Net-AR

∆RMSE (%): 0. Result ID: 699928f05698665ec490ca0c.

[Reporting paper](<https://www.sota2.com/research/paper/6971aefee25ad2fa9d1ed8d1.md>)

Recorded evaluation configuration: {"H":400}. This record does not assert independent reproduction.

### HINE-L2

∆RMSE (%): 34.7. Result ID: 699928f05698665ec490ca0b.

[Reporting paper](<https://www.sota2.com/research/paper/6971aefee25ad2fa9d1ed8d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79341e209131cea7c9df.md>)

Recorded evaluation configuration: {"H":400}. This record does not assert independent reproduction.

### MSR-HINE

∆RMSE (%): 62.8. Result ID: 699928f05698665ec490ca0a.

[Reporting paper](<https://www.sota2.com/research/paper/6971aefee25ad2fa9d1ed8d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6971aefee25ad2fa9d1ed8d1.md>)

Recorded evaluation configuration: {"H":400}. This record does not assert independent reproduction.

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