# Prompt Leakage Attack on FinanceBench

Source: [Prompt Leakage Attack on FinanceBench](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench>)

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

Data updated: 2026-02-26T05:24:06.168Z

Evaluation of model performance against prompt leakage attacks using the FinanceBench dataset, measured by metrics such as ASR500, ASR1000, ASR10000, w/l, and ARPT.

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 | Prompt Leakage Attack |
| Dataset (including any stated split) | FinanceBench |
| Metric | ASR500 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [ASR500](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench.md?metric=ASR500>): Higher is better; unit: PERCENT.

- [ASR1000](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench.md?metric=ASR1000>): Higher is better; unit: PERCENT.

- [ASR10000](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench.md?metric=ASR10000>): Higher is better; unit: PERCENT.

- [W/L Ratio](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench.md?metric=W%2FL%20Ratio>): Higher is better; unit: SCALAR.

- [ARPT](<https://www.sota2.com/research/sota/prompt-leakage-attack-on-financebench.md?metric=ARPT>): Higher is better; unit: SCALAR.

## Recorded results

| Method | ASR500 | Result date | Report paper |
| --- | --- | --- | --- |
| SFT | 88 | Not recorded | 699e81e66017c4460ffacd1b |
| OPTILEAK | 88 | Not recorded | 699e81e66017c4460ffacd1b |
| OPTILEAK | 86 | Not recorded | 699e81e66017c4460ffacd1b |
| OPTILEAK | 84 | Not recorded | 699e81e66017c4460ffacd1b |
| SFT | 78 | Not recorded | 699e81e66017c4460ffacd1b |
| SFT | 76 | Not recorded | 699e81e66017c4460ffacd1b |
| OPTILEAK | 64 | Not recorded | 699e81e66017c4460ffacd1b |
| SFT | 62 | Not recorded | 699e81e66017c4460ffacd1b |
| Base | 6 | Not recorded | 699e81e66017c4460ffacd1b |
| DPO | 6 | Not recorded | 699e81e66017c4460ffacd1b |
| Base | 4 | Not recorded | 699e81e66017c4460ffacd1b |
| DPO | 4 | Not recorded | 699e81e66017c4460ffacd1b |
| Base | 2 | Not recorded | 699e81e66017c4460ffacd1b |
| DPO | 2 | Not recorded | 699e81e66017c4460ffacd1b |
| Base | 0 | Not recorded | 699e81e66017c4460ffacd1b |
| DPO | 0 | Not recorded | 699e81e66017c4460ffacd1b |

### SFT

ASR500: 88. Result ID: 699eb0f25698665ec47e7b53.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-3B-Instruct"}. This record does not assert independent reproduction.

### OPTILEAK

ASR500: 88. Result ID: 699eb0f25698665ec47e7b55.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-3B-Instruct"}. This record does not assert independent reproduction.

### OPTILEAK

ASR500: 86. Result ID: 699eb0f25698665ec47e7b59.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

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

### OPTILEAK

ASR500: 84. Result ID: 699eb0f25698665ec47e7b5d.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-14B-Instruct"}. This record does not assert independent reproduction.

### SFT

ASR500: 78. Result ID: 699eb0f25698665ec47e7b5b.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-14B-Instruct"}. This record does not assert independent reproduction.

### SFT

ASR500: 76. Result ID: 699eb0f25698665ec47e7b57.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

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

### OPTILEAK

ASR500: 64. Result ID: 699eb0f25698665ec47e7b61.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### SFT

ASR500: 62. Result ID: 699eb0f25698665ec47e7b5f.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### Base

ASR500: 6. Result ID: 699eb0f25698665ec47e7b56.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

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

### DPO

ASR500: 6. Result ID: 699eb0f25698665ec47e7b58.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

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

### Base

ASR500: 4. Result ID: 699eb0f25698665ec47e7b5a.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-14B-Instruct"}. This record does not assert independent reproduction.

### DPO

ASR500: 4. Result ID: 699eb0f25698665ec47e7b5c.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-14B-Instruct"}. This record does not assert independent reproduction.

### Base

ASR500: 2. Result ID: 699eb0f25698665ec47e7b52.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-3B-Instruct"}. This record does not assert independent reproduction.

### DPO

ASR500: 2. Result ID: 699eb0f25698665ec47e7b54.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Model":"Qwen-2.5-3B-Instruct"}. This record does not assert independent reproduction.

### Base

ASR500: 0. Result ID: 699eb0f25698665ec47e7b5e.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### DPO

ASR500: 0. Result ID: 699eb0f25698665ec47e7b60.

[Reporting paper](<https://www.sota2.com/research/paper/699e81e66017c4460ffacd1b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74511e209131cea6105d.md>)

Recorded evaluation configuration: {"Model":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

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