Open-source GLiNER-PII scrubs 13 identifiers from a discharge note in 550ms on GPU
MaziyarPanahi · x · 2026-11-02
A hands-on test of open-source GLiNER-PII on a hospital discharge summary: on an L4 GPU it processed 9 lines in 1.45s (550ms on GPU), correctly tagging and removing 13 identifiers—names, MRNs, address, phone, dates, ZIP. But it missed half the hospital name ("General Hospital") and over-tagged non-identifiers like "day 3" and "within 7 days". The author asks clinicians whether they'd share a note after one pass like this, highlighting both the promise and limits of open PII de-identification for medical text.
More from Models
- SemiAnalysis: safety classifiers sandbag scores of Opus 5.5 and GPT-6 Astra — morqon · 2026-11-02
- Liquid AI's 1.7B LFM3 becomes first local model to nail live fixtures lookup on a MacBook — NielsRogge · 2026-11-02
- Claude upgrade-page traffic rebounded to 17.6M in September, near May's 18.6M peak — firstadopter · 2026-11-02
- OpenAI Dots user says bloat and broken infra drove him to try newcomer Muse — Koala_Confused · 2026-11-02
- Fable 5.5 reportedly one-shots animation in pure code as Anthropic's coding bet pays off — Hesamation · 2026-11-02
- One agent goal has been running for 4 days, burning 30K tokens — CtrlAltDwayne · 2026-11-02