OpenAI releases open-source weight model Privacy Filter for detecting and de-identifying PII in text

robot
Abstract generation in progress

Deep Tide TechFlow News: On April 22, according to official information, OpenAI released an open-source weighted model called Privacy Filter, which is used to detect and mask personal identity information (PII) in text. The model supports local operation and can complete long-text recognition and de-identification in a single forward pass, with up to 128,000 tokens of context. The Privacy Filter has a parameter size of 1.5 billion, with 50 million active parameters, and can identify sensitive information such as private names, addresses, email addresses, phone numbers, URLs, dates, account numbers and passwords, API keys, and more. OpenAI stated that the model is released under the Apache 2.0 license on Hugging Face and GitHub, and can be used for privacy-protecting processes such as training, indexing, logging, and auditing.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin