PII Prediction on SynthPAI (test)
84.8Accuracy (Age)AutoProfiler
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AutoProfilerLLM=GPT-52025.05 | 84.8 | 83.5 | 78.9 | 89.7 | 97.5 | 94.8 | 92.5 | 95.7 | |
| AutoProfilerLLM=GPT-42025.05 | 80.6 | 81 | 75.6 | 88.1 | 95.4 | 92 | 89.6 | 93.7 | |
| AutoProfilerLLM=Gemini-1.52025.05 | 77.8 | 71 | 71.1 | 83.1 | 88.6 | 88 | 82.2 | 85.6 | |
| AutoProfilerLLM=Claude-32025.05 | 75 | 75.5 | 68.9 | 92.5 | 90.8 | 84 | 87.5 | 92.8 | |
| AutoProfilerLLM=Qwen-22025.05 | 75 | 62 | 52 | 86.8 | 86.9 | 84 | 84.4 | 90.1 | |
| AutoProfilerLLM=Llama-32025.05 | 75 | 75 | 50 | 81.3 | 85.5 | 92 | 77.1 | 84.7 | |
| FTILLM=GPT-52025.05 | 71.9 | 75.2 | 68.4 | 84.5 | 78.2 | 89.4 | 81.5 | 93.2 | |
| PIELLM=GPT-52025.05 | 71.9 | 75.8 | 67.6 | 84.5 | 80.6 | 89.4 | 82.6 | 93.2 | |
| PIELLM=Llama-32025.05 | 70.1 | 73.8 | 46.7 | 81.2 | 73.3 | 85.8 | 72.9 | 82.6 | |
| PIELLM=GPT-42025.05 | 69.8 | 73.2 | 67.5 | 81.3 | 74.2 | 88.9 | 79.8 | 92.8 | |
| FTILLM=GPT-42025.05 | 69.4 | 73 | 66.7 | 80 | 73.9 | 88 | 79.2 | 92.8 | |
| FTILLM=Llama-32025.05 | 69.4 | 73 | 46.7 | 80.6 | 72.9 | 84 | 72.9 | 82 | |
| PIELLM=Gemini-1.52025.05 | 67.4 | 54.6 | 52.5 | 67.6 | 66.3 | 84.6 | 79.3 | 77.2 | |
| FTILLM=Gemini-1.52025.05 | 66.7 | 53.5 | 51.1 | 66.3 | 65.7 | 84 | 78.1 | 76.6 | |
| PIELLM=Qwen-22025.05 | 52.9 | 60.8 | 42.3 | 76.9 | 72.6 | 81.5 | 73.6 | 89 | |
| FTILLM=Qwen-22025.05 | 50 | 59 | 40 | 76.9 | 71.1 | 80 | 72.9 | 88.3 | |
| PIELLM=Claude-32025.05 | 47.9 | 70.1 | 64.5 | 72.8 | 76 | 79.5 | 87.2 | 91.9 | |
| FTILLM=Claude-32025.05 | 47.2 | 69 | 64.5 | 71.2 | 75.4 | 78 | 86.5 | 91.9 | |
| PII-ExtractorLLM=-2025.05 | 21.2 | 41.3 | 26.2 | 29.5 | 40.7 | 62.8 | 48.2 | 67.4 |