Multilingual PII Detection and Redaction with LFM2.5 Encoder

This demo shows how Liquid AI's LFM2.5 Encoder detects and redacts 40 types of personally identifiable information across 16 languages. The encoder processes the complete sequence bidirectionally, combining contextual understanding with structured entity labels. It can identify names, addresses, financial information, government identifiers, device data, GPS coordinates, cryptocurrency wallets, API keys, and other sensitive fields. The model can run inside an organization's own pipeline, allowing raw personal data to remain within its environment. It also avoids token-by-token generation and generated-output parsing, making the approach suitable for deterministic preprocessing systems. 🔗 Links: • Hugging Face demo: https://huggingface.co/spaces/LiquidAI/pii-detection • Blog post: https://www.liquid.ai/blog/lfm2-5-encoders Connect with Liquid AI: • Careers: https://www.liquid.ai/careers • Hugging Face: https://huggingface.co/LiquidAI • Discord: https://discord.com/invite/liquid-ai • X: https://x.com/LiquidAI • LinkedIn: https://www.linkedin.com/company/liquid-ai-inc/ • GitHub: https://github.com/Liquid4All/cookbook • Substack: https://liquidai.substack.com/