# Financial Question Answering on ConvFinQA (test)

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

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

Data updated: 2026-05-08T07:37:06.945Z

Evaluation of model performance on the test split of the ConvFinQA dataset for Financial Question Answering, measured by Execution Accuracy and Turn Accuracy.

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) | ConvFinQA (test) |
| Metric | Execution Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

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

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

## Recorded results

| Method | Execution Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| FinAgent-RAG | 78.46 | Not recorded | 69fd6de95497b3b670c892a9 |
| IterRAG | 69.14 | Not recorded | 69fd6de95497b3b670c892a9 |
| CRAG | 66.72 | Not recorded | 69fd6de95497b3b670c892a9 |
| Self-RAG | 64.83 | Not recorded | 69fd6de95497b3b670c892a9 |
| ReAct | 62.47 | Not recorded | 69fd6de95497b3b670c892a9 |
| Advanced RAG | 60.34 | Not recorded | 69fd6de95497b3b670c892a9 |
| FinMA | 57.92 | Not recorded | 69fd6de95497b3b670c892a9 |
| Naive RAG | 56.14 | Not recorded | 69fd6de95497b3b670c892a9 |
| Zero-shot LLM | 51.78 | Not recorded | 69fd6de95497b3b670c892a9 |

### FinAgent-RAG

Execution Accuracy: 78.46. Result ID: 69fd929d47f34ce924e03fd6.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### IterRAG

Execution Accuracy: 69.14. Result ID: 69fd929d47f34ce924e03fd5.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70f21e209131cea4e48a.md>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### CRAG

Execution Accuracy: 66.72. Result ID: 69fd929d47f34ce924e03fd4.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75111e209131cea65797.md>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### Self-RAG

Execution Accuracy: 64.83. Result ID: 69fd929d47f34ce924e03fd3.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70e51e209131cea4dc1c.md>)

[Implementation](<https://selfrag.github.io/>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### ReAct

Execution Accuracy: 62.47. Result ID: 69fd929d47f34ce924e03fd2.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eb41e209131cea41992.md>)

[Implementation](<https://react-lm.github.io/>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### Advanced RAG

Execution Accuracy: 60.34. Result ID: 69fd929d47f34ce924e03fd1.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### FinMA

Execution Accuracy: 57.92. Result ID: 69fd929d47f34ce924e03fcf.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471fe.md>)

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

### Naive RAG

Execution Accuracy: 56.14. Result ID: 69fd929d47f34ce924e03fd0.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b91e209131cea23c26.md>)

[Implementation](<https://github.com/huggingface/transformers/blob/master/examples/rag/>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

### Zero-shot LLM

Execution Accuracy: 51.78. Result ID: 69fd929d47f34ce924e03fce.

[Reporting paper](<https://www.sota2.com/research/paper/69fd6de95497b3b670c892a9.md>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-V3"}. This record does not assert independent reproduction.

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