Data useful for language model (LM) training can contain private or unwanted information that is at risk of being learned or memorized. We propose Surprise Masking (SM), a simple form of perplexity filtering that uses a publicly available reference model to identify unwanted spans and suppress learning from them while...
Riya Tyagi, Sahra Ghalebikesabi, Vinnie Monaco et al.· 0 citations
Methods for training language models on sensitive data differ in how they balance capability gains against the risk of leaking private information. In this paper, we present a controlled comparison of data-level and training-level interventions for navigating the capability-privacy tradeoff on synthetic medical records...
Zihao Zhao, Avi Schwarzschild, Casey Meehan et al.· 0 citations
OpenAI Privacy Filter is a compact, bidirectional token-classification model for detecting and redacting personally identifiable information (PII) and secrets in unstructured text. The model is derived from an autoregressively pretrained checkpoint and converted into a bidirectional, banded-attention classifier that la...
Charles de Bourcy, Sahra Ghalebikesabi, Avi Schwarzschild et al.· 2 citations· ⚡1
This work introduces A 2 -Judger, a novel MLLM-based A gentic instantiation of A uto Judger equipped with semantic-aware retrieval and dynamic memory that significantly improves sample efficiency while maintaining reliable evaluation results.
Xuanwen Ding, Chengjun Pan, Zejun Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
