# Long-document Question Answering on FinanceBench (FB)

Source: [Long-document Question Answering on FinanceBench (FB)](<https://www.sota2.com/research/sota/long-document-question-answering-on-financebench-fb>)

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

Data updated: 2026-04-27T07:09:40.750Z

Evaluation of Long-document Question Answering performance on the FinanceBench (FB) dataset, measured using 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 | Long-document Question Answering |
| Dataset (including any stated split) | FinanceBench (FB) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/long-document-question-answering-on-financebench-fb.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| SLIDERS | 89.33 | Not recorded | 69eeed9955a3c32af89f97f5 |
| Basemodel | 84.67 | Not recorded | 69eeed9955a3c32af89f97f5 |
| SLIDERS | 82.1 | Not recorded | 69eeed9955a3c32af89f97f5 |
| Basemodel | 82 | Not recorded | 69eeed9955a3c32af89f97f5 |
| GraphRAG | 75.33 | Not recorded | 69eeed9955a3c32af89f97f5 |
| RLM | 75.33 | Not recorded | 69eeed9955a3c32af89f97f5 |
| LongRAG | 72 | Not recorded | 69eeed9955a3c32af89f97f5 |
| Chain of Agent | 71.3 | Not recorded | 69eeed9955a3c32af89f97f5 |
| DocETL | 63.33 | Not recorded | 69eeed9955a3c32af89f97f5 |
| RAG | 62.67 | Not recorded | 69eeed9955a3c32af89f97f5 |

### SLIDERS

Accuracy: 89.33. Result ID: 69ef0a7c47f34ce924d4bc6d.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Implementation](<https://github.com/stanford-oval/sliders>)

Recorded evaluation configuration: {"LLMs":"GPT 4.1 & GPT 4.1-mini"}. This record does not assert independent reproduction.

### Basemodel

Accuracy: 84.67. Result ID: 69ef0a7c47f34ce924d4bc69.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

Recorded evaluation configuration: {"LLMs":"Qwen3.5 122B-A10B"}. This record does not assert independent reproduction.

### SLIDERS

Accuracy: 82.1. Result ID: 69ef0a7c47f34ce924d4bc6e.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Implementation](<https://github.com/stanford-oval/sliders>)

Recorded evaluation configuration: {"LLMs":"Qwen3.5 122B-A10B"}. This record does not assert independent reproduction.

### Basemodel

Accuracy: 82. Result ID: 69ef0a7c47f34ce924d4bc68.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

Recorded evaluation configuration: {"LLMs":"GPT 4.1"}. This record does not assert independent reproduction.

### GraphRAG

Accuracy: 75.33. Result ID: 69ef0a7c47f34ce924d4bc67.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa770d1e209131cea70c5a.md>)

Recorded evaluation configuration: {"LLMs":"Qwen3-4B & GPT 4.1"}. This record does not assert independent reproduction.

### RLM

Accuracy: 75.33. Result ID: 69ef0a7c47f34ce924d4bc6c.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6955ff9153fdd5c36173e066.md>)

[Implementation](<https://github.com/alexzhang13/rlm/tree/main>)

Recorded evaluation configuration: {"LLMs":"GPT 5 & GPT 5-mini"}. This record does not assert independent reproduction.

### LongRAG

Accuracy: 72. Result ID: 69ef0a7c47f34ce924d4bc66.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea63323.md>)

Recorded evaluation configuration: {"LLMs":"Qwen3-4B & GPT 4.1"}. This record does not assert independent reproduction.

### Chain of Agent

Accuracy: 71.3. Result ID: 69ef0a7c47f34ce924d4bc6b.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73721e209131cea5c8b1.md>)

Recorded evaluation configuration: {"LLMs":"GPT 5 & GPT 5-mini"}. This record does not assert independent reproduction.

### DocETL

Accuracy: 63.33. Result ID: 69ef0a7c47f34ce924d4bc6a.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77cb1e209131cea75614.md>)

Recorded evaluation configuration: {"LLMs":"GPT 4.1"}. This record does not assert independent reproduction.

### RAG

Accuracy: 62.67. Result ID: 69ef0a7c47f34ce924d4bc65.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9955a3c32af89f97f5.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: {"LLMs":"Qwen3-4B & GPT 4.1"}. This record does not assert independent reproduction.

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