List your company
ResearchFormerly WizWand

Privacy Extraction on PersonaMem v2 (test)

0.9448F1 ScoreMemPrivacy-4B-RL
0.0341760.2705880.5070.743412May 10, 2026
Updated 6mo ago

Evaluation results

MethodLinks
MemPrivacy-4B-RLModel Category=Privacy Models, Model Scale=4B, Training Stage=RL2026.050.94480.94540.95110.44
MemPrivacy-4B-SFTModel Category=Privacy Models, Model Scale=4B, Training Stage=SFT2026.050.94430.94520.950.67
MemPrivacy-1.7B-RLModel Category=Privacy Models, Model Scale=1.7B, Training Stage=RL2026.050.94020.94390.9420.48
MemPrivacy-0.6B-RLModel Category=Privacy Models, Model Scale=0.6B, Training Stage=RL2026.050.9340.94070.93310.43
MemPrivacy-1.7B-SFTModel Category=Privacy Models, Model Scale=1.7B, Training Stage=SFT2026.050.93160.93650.93210.59
DeepSeek-V3.2-ThinkModel Category=General Models2026.050.92180.92160.928439.6
MemPrivacy-0.6B-SFTModel Category=Privacy Models, Model Scale=0.6B, Training Stage=SFT2026.050.92080.92780.91930.7
GPT-5.2Model Category=General Models2026.050.88060.8770.89483.23
GLM-5Model Category=General Models2026.050.87870.87970.88278.56
Kimi-K2Model Category=General Models2026.050.87350.88390.86941.81
LongCat-Flash-ChatModel Category=General Models2026.050.86790.87160.86911.75
DeepSeek-V3.2Model Category=General Models2026.050.86610.86930.86563.55
Gemini-3.1-ProModel Category=General Models2026.050.86590.86680.868229.21
Qwen3-4BModel Category=General Models, Model Scale=4B2026.050.86360.86380.86990.64
DeepSeek-V3.1-TerminusModel Category=General Models2026.050.85910.86230.85841.68
OpenAI-Privacy-FilterModel Category=Privacy Models, Type=Token-classification2026.050.85270.8580.85750.24
Qwen3-0.6BModel Category=General Models, Model Scale=0.6B2026.050.09320.08050.13492.36
Qwen3-1.7BModel Category=General Models, Model Scale=1.7B2026.050.06920.05970.10532.17