rocky/OSGKeyboard
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(keyboard): expand contextual skills and managed flows

Browse Source 
Add local clipboard intent recommendations, webpage and phone actions, and safer host handoffs. Refine managed gateway, catalog refresh, onboarding, and adaptive polish behavior.
This commit is contained in:
Rocky
2026-08-23 14:02:33 +08:00
parent 7df5bbdaa0
commit 2cc81c4628
68 changed files with 10368 additions and 7765 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ModelTraining/ClipboardSemantics/README.md
+17 -14
View File
@@ -5,17 +5,19 @@ for OSGKeyboard's fully local clipboard analyzer.
## Scope
The model suite predicts four independent intents (`task`, `question`,
`invitation`, and `complaint`) plus three-way sentiment. Apple data detectors
remain responsible for dates, addresses, phone numbers, and URLs; `NLTagger`
provides best-effort person and organization names.
The model suite predicts five independent intents (`task`, `question`,
`invitation`, `complaint`, and `replyableMessage`) plus three-way sentiment.
`replyableMessage` distinguishes messages that invite a response from terminal
acknowledgments, personal notes, quoted questions, and factual notices. Apple
data detectors remain responsible for dates, addresses, phone numbers, and
URLs; `NLTagger` provides best-effort person and organization names.
The corpus contains 6,334 Chinese and English records:
The corpus contains 7,272 Chinese and English records:
- 4,100 generated training records
- 1,080 generated validation records
- 1,080 template-held-out test records
- 74 manually authored golden records
- 4,660 generated training records
- 1,260 generated validation records
- 1,260 template-held-out test records
- 92 manually authored golden records
No user clipboard content is included.
@@ -38,11 +40,12 @@ transfer models depend on `NLContextualEmbedding` assets that are not guaranteed
to exist in a simulator or keyboard-extension runtime, so they remain evaluation
candidates only.
The selected task, question, and invitation models passed the automatic-routing
precision gates. The complaint model is packaged for further evaluation but its
automatic-routing flag remains disabled because golden-set precision is 88.89%,
below the 90% release gate. Sentiment returns `unknown` unless confidence and
top-two margin checks both pass.
The selected intent models passed the automatic-routing precision gates. The
replyable-message model reached 100% test precision and 97.96% golden precision;
its measured recall remains part of release monitoring. The complaint model is
also high precision but remains conservative because golden recall is limited.
Sentiment returns `unknown` unless confidence and top-two margin checks both
pass.
Synthetic results are not treated as production truth. Real opt-in, anonymized
or manually reviewed examples are still required before widening labels or