Skip to content

Releases

v0.4.0

Choose a tag to compare

@gabrielkpo gabrielkpo released this 17 Nov 10:01
7b750b5

0.4.0 (2026-11-16)

Features

  • bridge: a container image for alias-pii serve (#80) (eee5eee)
  • bridge: a local token when the bridge holds the Mistral key (#79) (5618255)
  • bridge: accept streamed requests, restored whole then sent as events (#76) (82c5e44)
  • bridge: local Mistral-compatible endpoint that masks and restores (#58) (0346204)
  • bridge: measure the bridge against the real Mistral API (#81) (63282d8)
  • cli: alias-pii serve, the Mistral key in the OS keychain (#77) (c0f55d2)
  • desktop: a first-launch setup and a clearer window (#89) (a9a3434)
  • desktop: Alias as a macOS app (Tauri), built and launched in CI (#86) (9405782)
  • desktop: the app's own blue instead of orange (#90) (69dd17c)
  • release: attach the macOS app to each release (#87) (a1c767b)

Bug Fixes

  • bridge: forward nothing unmasked, never restore values into links (#60) (3b44f63)
  • bridge: gateway errors, body cap, answers in parts (#65) (d341a5e)
  • bridge: give every occurrence of an aliased value the same alias (#62) (4bd07d9)
  • bridge: keep Mistral's gzip answers intact in the measure (#83) (3217c9d)
  • bridge: refuse to start without a model unless --rules-only (#61) (f99e3fa)
  • bridge: run the measure in one event loop (#82) (30ca330)
  • desktop: sign the app without the hardened runtime so its bridge starts (#88) (9ab431d)
  • desktop: the app alone uses the keychain, so macOS stops asking (#91) (322a0ca)
  • masking: keep e-mails and identifiers apart when a dot or dash differs (#64) (77b7ab7)
  • masking: one alias for a UK or US phone found by the model (#72) (f83a218)
  • masking: restore aliases an LLM rewrote in another case or form (#75) (08be5ec)
  • masking: take values from the text, keep the masked text out of repr (#70) (b81edb6)
  • rules: read a file number when one word qualifies the keyword (#71) (83735b4)
  • rules: read through typographic spaces and invisible characters (#63) (97fb597)

Documentation

  • bridge: record the first measure against the real Mistral API (#84) (60b4f9a)
  • bring the README up to date with main (#68) (ce00de3)
  • refresh stale docs and rewrite working notes as neutral records (#69) (7249505)

v0.3.0

Choose a tag to compare

@gabrielkpo gabrielkpo released this 11 Nov 19:53
472a6cb

0.3.0 (2026-11-11)

Features

  • onnx: ONNX export and detector (the demo's model) (#37) (79fbeff)

Bug Fixes

  • deps: transformers 5 (>=5.10), closing its security alerts (#54) (fb34fe2)

Documentation

  • readme: Mistral zero-shot baseline results (#39) (1efe31e)

v0.2.0

Choose a tag to compare

@gabrielkpo gabrielkpo released this 11 Nov 17:33
0cb401f

0.2.0 (2026-11-11)

Features

  • app: local FastAPI to mask a pasted text (#31) (7122301)
  • app: web page to paste a text and copy it masked (#32) (fe7ad6f)
  • benchmark: harness with rules, Presidio and GLiNER baselines (step 3) (#12) (e85a387)
  • benchmark: hybrid rules + GLiNER and Mistral zero-shot baselines (#14) (162b7b7)
  • dataset: grid generator, balancer, build command and datasheet (dataset v0.2, part 2) (#10) (c617225)
  • dataset: identity factory and quality gates (dataset v0.2, part 1) (#9) (720adaf)
  • dataset: LLM-written documents and public corpora (dataset v0.3 and v0.4) (#11) (4021ca5)
  • demo: Hugging Face Space and model card, not published (#35) (8eaabf2)
  • demo: put the values back in the LLM's answer (#48) (e3c3eb3)
  • demo: run the detector in the browser on a static Hugging Face Space (#41) (7a8ac62)
  • demo: tell first visitors the model takes about 5 minutes to download (#47) (04e39ae)
  • detector: add identifiers announced by a person's-file keyword (#45) (ea952d9)
  • detector: rules + model detector, step 5 (#30) (28f25d1)
  • docker: CPU image of the API and page, tested on Actions (#33) (d945e98)
  • eval: model choice for step 4, CPU latency and careful int8 modes (#19) (52a4674)
  • open-sets: TAB as a realistic English test set (#46) (072f8c8)
  • readers: read .txt, .md, .docx and text PDFs into plain text (#5) (067acab)
  • rules: FR + EN rule detector, stable alias masking and restoring (#1) (b05e91b)
  • synthetic: generate annotated French and English documents (#7) (fa624b7)
  • train: fine-tuning command for the three candidates, heavy runs on GitHub CPU machines (#15) (2e3cf94)

Bug Fixes

  • ci: compare index.html without what Hugging Face adds when serving it (#50) (99f533c)
  • dataset: LLM writer prompt v4 and two new gates, after a first run with the API (#18) (92d4c82)
  • demo: allow the runtime's blob module in the CSP (#44) (c1f9624)
  • release: tag as vX.Y.Z and align release-please config with secours-tech (#4) (1f35b27)
  • train: gradient accumulation option, so mDeBERTa fits a T4 with the same batch (#17) (6eacd2d)
  • train: turn off ModernBERT's compiled MLP, which crashes training on Kaggle GPUs (#16) (fdfe69d)

Documentation

  • dataset: specify how the training data is built, labelled and checked (#8) (f5ea711)
  • open-sets: TAB test results, read once (#51) (d3e5229)
  • README as a project page with architecture, roadmap and evaluation plan (#6) (f02fb7a)
  • readme: status banner reflects the trained model and its measured results (#29) (5eaea35)