# Financial Question Answering on FinanceBench (test)

Source: [Financial Question Answering on FinanceBench (test)](<https://www.sota2.com/research/sota/financial-question-answering-on-financebench-test>)

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

Data updated: 2026-02-26T06:00:44.341Z

Evaluation of Financial Question Answering performance on the test split of the FinanceBench dataset, using ROUGE-L and Numeric Match 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 | Financial Question Answering |
| Dataset (including any stated split) | FinanceBench (test) |
| Metric | ROUGE-L |
| 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

- [ROUGE-L](<https://www.sota2.com/research/sota/financial-question-answering-on-financebench-test.md?metric=ROUGE-L>): Higher is better; unit: PERCENT.

- [Numeric Match](<https://www.sota2.com/research/sota/financial-question-answering-on-financebench-test.md?metric=Numeric%20Match>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ROUGE-L | Result date | Report paper |
| --- | --- | --- | --- |
| Oracle Page | 20 | Not recorded | 699bded809c33900fcd80310 |
| Oracle Doc | 16 | Not recorded | 699bded809c33900fcd80310 |
| Learned Page Scorer | 15 | Not recorded | 699bded809c33900fcd80310 |
| BGE-M3 + Multi-HyDE + ReRanker | 7 | Not recorded | 699bded809c33900fcd80310 |

### Oracle Page

ROUGE-L: 20. Result ID: 699d068f5698665ec4a6a961.

[Reporting paper](<https://www.sota2.com/research/paper/699bded809c33900fcd80310.md>)

Recorded evaluation configuration: {"Generator Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### Oracle Doc

ROUGE-L: 16. Result ID: 699d068f5698665ec4a6a960.

[Reporting paper](<https://www.sota2.com/research/paper/699bded809c33900fcd80310.md>)

Recorded evaluation configuration: {"Generator Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### Learned Page Scorer

ROUGE-L: 15. Result ID: 699d068f5698665ec4a6a95f.

[Reporting paper](<https://www.sota2.com/research/paper/699bded809c33900fcd80310.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699bded809c33900fcd80310.md>)

Recorded evaluation configuration: {"Generator Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### BGE-M3 + Multi-HyDE + ReRanker

ROUGE-L: 7. Result ID: 699d068f5698665ec4a6a95e.

[Reporting paper](<https://www.sota2.com/research/paper/699bded809c33900fcd80310.md>)

Recorded evaluation configuration: {"Generator Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

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