Skip to content

docs(WO88): the backbone lever result #373

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ronaldtse merged 1 commit into from
Oct 27, 2026
Merged

docs(WO88): the backbone lever result #373

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions TODO.sota/88-secryst-migration.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,3 +40,19 @@ backbone (the legacy may be base-class), corpus enrichment (merge the
Chuon Nath IPA + SBBIC vocab layers), longer schedule. The migration
contract holds: the framework path produces a honest, reproducible
28.99 where the legacy number was undocumented.


## Backbone lever RESULT (2026-10-27): ByT5-base 28.81% best (ep2)

Like-for-like standings on the secryst test split:
- Legacy artifact: **27.57%**
- Framework ByT5-small (8ep): 28.99%
- Framework ByT5-base (4ep): **28.81%** (best; ep3 28.92)

The backbone lever is near-exhausted — the remaining ~1.2-point gap is
the CORPUS: the legacy's undocumented training set evidently exceeds
our 17k pairs. The recorded next lever is corpus enrichment (merge the
Chuon Nath IPA layer + SBBIC vocabulary + the khm-latn pairs), which
is the WO68 data work. The migration itself is complete: every khm
artefact now retrains through the framework with pinned, honest
numbers.
Loading