# time series feature understanding benchmark

Source: [time series feature understanding benchmark](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark>)

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

Data updated: 2026-04-08T07:18:32.790Z

Evaluation of model performance on the time series feature understanding benchmark, focusing on the decomposition and measurement of underlying time series characteristics such as Trend, Seasonality, and Anomalies.

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 | time series feature understanding |
| Dataset (including any stated split) | time series feature understanding benchmark |
| Metric | Trend Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Trend Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Trend%20Score>): Higher is better; unit: PERCENT.

- [Seasonality Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Seasonality%20Score>): Higher is better; unit: PERCENT.

- [Anomaly Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Anomaly%20Score>): Higher is better; unit: PERCENT.

- [Volatility Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Volatility%20Score>): Higher is better; unit: PERCENT.

- [Structural Break Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Structural%20Break%20Score>): Higher is better; unit: PERCENT.

- [Fixed Correlation Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Fixed%20Correlation%20Score>): Higher is better; unit: PERCENT.

- [Lagged Correlation Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Lagged%20Correlation%20Score>): Higher is better; unit: PERCENT.

- [Changing Correlation Score](<https://www.sota2.com/research/sota/time-series-feature-understanding-benchmark.md?metric=Changing%20Correlation%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Trend Score | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4o-mini | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| GPT-4o | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| Gemini-2.5 | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| Gemini-2.5 | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| ChatTS | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| TS-Agent | 98 | Not recorded | 68fa7c041e209131cea8be3d |
| Mistral-7B | 67 | Not recorded | 68fa7c041e209131cea8be3d |
| Phi-3 | 48 | Not recorded | 68fa7c041e209131cea8be3d |

### GPT-4o-mini

Trend Score: 98. Result ID: 69d6014047f34ce924be27a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Text"}. This record does not assert independent reproduction.

### GPT-4o

Trend Score: 98. Result ID: 69d6014047f34ce924be27a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Text"}. This record does not assert independent reproduction.

### Gemini-2.5

Trend Score: 98. Result ID: 69d6014047f34ce924be27a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Text"}. This record does not assert independent reproduction.

### Gemini-2.5

Trend Score: 98. Result ID: 69d6014047f34ce924be27a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Vision"}. This record does not assert independent reproduction.

### ChatTS

Trend Score: 98. Result ID: 69d6014147f34ce924be27a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78041e209131cea76582.md>)

Recorded evaluation configuration: {"Modality":"TS"}. This record does not assert independent reproduction.

### TS-Agent

Trend Score: 98. Result ID: 69d6014147f34ce924be27a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Mistral-7B

Trend Score: 67. Result ID: 69d6014047f34ce924be27a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Text"}. This record does not assert independent reproduction.

### Phi-3

Trend Score: 48. Result ID: 69d6014147f34ce924be27a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c041e209131cea8be3d.md>)

Recorded evaluation configuration: {"Modality":"Text"}. This record does not assert independent reproduction.

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