# Personal Information Extraction on Synthetic Personal Profile Dataset

Source: [Personal Information Extraction on Synthetic Personal Profile Dataset](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset>)

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

Data updated: 2026-04-08T09:07:33.522Z

Evaluation of Personal Information Extraction (PIE) performance on the Synthetic Personal Profile Dataset. Metrics include Accuracy, Rouge-1, and BERT Score calculated across specific entities such as Name, Email, Phone, Work Experience, and Affiliation.

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 | Personal Information Extraction |
| Dataset (including any stated split) | Synthetic Personal Profile Dataset |
| Metric | Accuracy (Email) |
| 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 (Email)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Email)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Email)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Email)>): Higher is better; unit: SCALAR.

- [Accuracy (Telephone)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Telephone)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Telephone)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Telephone)>): Higher is better; unit: SCALAR.

- [Accuracy (Mail Address)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Mail%20Address)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Mail Address)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Mail%20Address)>): Higher is better; unit: SCALAR.

- [Accuracy (Name)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Name)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Name)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Name)>): Higher is better; unit: SCALAR.

- [Accuracy (Work Experience)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Work%20Experience)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Work Experience)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Work%20Experience)>): Higher is better; unit: SCALAR.

- [Accuracy (Education Experience)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Education%20Experience)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Education Experience)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Education%20Experience)>): Higher is better; unit: SCALAR.

- [Accuracy (Affiliation)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Affiliation)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Affiliation)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Affiliation)>): Higher is better; unit: SCALAR.

- [Accuracy (Occupation)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=Accuracy%20(Occupation)>): Higher is better; unit: PERCENT.

- [ROUGE-1/BERT Score (Occupation)](<https://www.sota2.com/research/sota/personal-information-extraction-on-synthetic-personal-profile-dataset.md?metric=ROUGE-1%2FBERT%20Score%20(Occupation)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy (Email) | Result date | Report paper |
| --- | --- | --- | --- |
| Prompt Injection Defense | 54 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 5 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 5 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 3 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 1 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 1 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 0 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 0 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 0 | Not recorded | 68fa76c31e209131cea6ed51 |
| Prompt Injection Defense | 0 | Not recorded | 68fa76c31e209131cea6ed51 |

### Prompt Injection Defense

Accuracy (Email): 54. Result ID: 69d61a9247f34ce924beda2e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"flan-ul2"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 5. Result ID: 69d61a9247f34ce924beda29.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"gpt-3.5-turbo"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 5. Result ID: 69d61a9247f34ce924beda2d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"internlm-chat-7b"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 3. Result ID: 69d61a9247f34ce924beda2b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"vicuna-7b-v1.3"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 1. Result ID: 69d61a9247f34ce924beda27.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"gemini-pro"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 1. Result ID: 69d61a9247f34ce924beda2c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"llama-2-7b-chat-hf"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 0. Result ID: 69d61a9247f34ce924beda25.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"text-bison-001"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 0. Result ID: 69d61a9247f34ce924beda26.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"chat-bison-001"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 0. Result ID: 69d61a9247f34ce924beda28.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"gpt-4"}. This record does not assert independent reproduction.

### Prompt Injection Defense

Accuracy (Email): 0. Result ID: 69d61a9247f34ce924beda2a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76c31e209131cea6ed51.md>)

Recorded evaluation configuration: {"LLM":"vicuna-13b-v1.3"}. This record does not assert independent reproduction.

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