# Evidence Retrieval on TrendFact (test)

Source: [Evidence Retrieval on TrendFact (test)](<https://www.sota2.com/research/sota/evidence-retrieval-on-trendfact-test>)

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

Data updated: 2026-07-07T08:31:13.008Z

Evaluation of model performance on the test split of the TrendFact dataset for Evidence Retrieval, measured using ranking metrics like R@K.

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 | Evidence Retrieval |
| Dataset (including any stated split) | TrendFact (test) |
| Metric | R@1 |
| Unit | PERCENT |
| 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

- [R@1](<https://www.sota2.com/research/sota/evidence-retrieval-on-trendfact-test.md?metric=R%401>): Higher is better; unit: PERCENT.

- [R@2](<https://www.sota2.com/research/sota/evidence-retrieval-on-trendfact-test.md?metric=R%402>): Higher is better; unit: PERCENT.

- [R@3](<https://www.sota2.com/research/sota/evidence-retrieval-on-trendfact-test.md?metric=R%403>): Higher is better; unit: PERCENT.

- [R@5](<https://www.sota2.com/research/sota/evidence-retrieval-on-trendfact-test.md?metric=R%405>): Higher is better; unit: PERCENT.

## Recorded results

| Method | R@1 | Result date | Report paper |
| --- | --- | --- | --- |
| bge-m3(dense) | 14.1 | Not recorded | 68fa78b41e209131cea79de6 |
| BM25 | 12.99 | Not recorded | 68fa78b41e209131cea79de6 |
| text-emb-ada-002 | 12.1 | Not recorded | 68fa78b41e209131cea79de6 |
| BM25 -w/o date | 12.08 | Not recorded | 68fa78b41e209131cea79de6 |

### bge-m3(dense)

R@1: 14.1. Result ID: 6a4cb93660b88bb339c88eff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78b41e209131cea79de6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73d91e209131cea5e7ab.md>)

[Implementation](<https://github.com/FlagOpen/FlagEmbedding>)

Recorded evaluation configuration: {"mode":"dense"}. This record does not assert independent reproduction.

### BM25

R@1: 12.99. Result ID: 6a4cb93660b88bb339c88efd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78b41e209131cea79de6.md>)

Recorded evaluation configuration: {"with date":"true"}. This record does not assert independent reproduction.

### text-emb-ada-002

R@1: 12.1. Result ID: 6a4cb93660b88bb339c88efe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78b41e209131cea79de6.md>)

Recorded evaluation configuration: {"provider":"OpenAI"}. This record does not assert independent reproduction.

### BM25 -w/o date

R@1: 12.08. Result ID: 6a4cb93660b88bb339c88efc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa78b41e209131cea79de6.md>)

Recorded evaluation configuration: {"with date":"false"}. This record does not assert independent reproduction.

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