PII Detection on PII-Hard v1 (test)
0.893Entity F1 ScoreDeepSeekV3
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DeepSeekV3Model Category=API-based Large Language Model2025.02 | 0.893 | 0.838 | 0.838 | |
| Llama3.1Model Category=Open-source Large Language Model2025.02 | 0.893 | 0.84 | 0.841 | |
| Qwen2.5Model Category=Open-source Large Language Model2025.02 | 0.876 | 0.804 | 0.806 | |
| GPT4oModel Category=API-based Large Language Model2025.02 | 0.869 | 0.817 | 0.819 | |
| Claude3.5Model Category=API-based Large Language Model2025.02 | 0.857 | 0.813 | 0.818 | |
| Qwen2.5-SLMModel Category=Open-source Small Language Model2025.02 | 0.81 | 0.591 | 0.594 | |
| Llama3.1-SLMModel Category=Open-source Small Language Model2025.02 | 0.798 | 0.718 | 0.722 | |
| BiLSTM-CRFModel Category=Traditional Model2025.02 | 0.684 | — | — |