Skip to content

data(WO87): evidence-based rename map + defect register #360

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 24, 2026
Merged

data(WO87): evidence-based rename map + defect register #360

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,7 +88,7 @@ CER unless stated).
| model | task | size | headline metric |
|---|---|---|---|
| `ara-diac-2.0` | Arabic diacritization | 2.78 GB fp32 | SadeedDiac-25 DER **2.29%** (beats GLM-5.2 2.69); WikiNews-2024 17.38 WER |
| `ara-diac-news-1.1` | Arabic news register | 2.16 GB fp32 | WikiNews multiref 8.85 WER |
| ~~`ara-diac-news-1.1`~~ | Arabic news register | — | **WITHDRAWN** — artifact lost (never shipped; WO87). `ara-diac-2.0` covers the register |
| `ara-diac-small-2.1` | Arabic, fast tier | 1.42 GB fp32 | fullset DER 4.57 (teacher 2.29) |
| `ara-diac-plane-large-1.0` | Arabic, int8 fast tier | 0.87 GB int8 | DER 2.74 / WER 8.61 |
| `urd-diac-base-1.0` | Urdu diacritization | 1.35 GB fp16 | CER **3.57%** |
Expand Down
29 changes: 29 additions & 0 deletions docs/naming/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
# Target documentation — WO87 evidence (2026-10-24)

`target-documentation.json`: every index id decoded through the
reference loader with a per-language probe — the evidence for
target-accurate naming.

## Verified target families

| family | ids | output sample |
|---|---|---|
| Arab-vocalized (diac) | ara-diac-*, urd-diac-* | 'السَّلَامُ عَلَيْكُمْ' |
| Heb-vocalized (nikud) | heb-diac-* (except plane: load ERROR) | 'שָׁלוֹם' |
| Thai-phonemic (toned, dots) | tha-g2p-* , tha-phonemic-1.0 | 'sa˨˩.wat̚˨˩.diː˧' |
| IPA (narrow, stress) | urd-g2p-* | 'ɪst̪ɪqˈbaːl' |
| Latin-letter phonemic | fas-g2p-1.0 | 'salAm' |
| Latin romanization | khm-latn-1.0 | 'suostei' |

## Defects found by the sweep

1. **ara-diac-news-1.1: artifact LOST** — index/release 404; bytes
exist nowhere (bucket zip is news-1.0's, sha differs). Withdrawn in
models.yaml; restore = re-export from run-029 best as a new version.
2. **Plane family non-contract** (ara-diac-plane-1.0,
ara-diac-plane-large-1.0, heb-diac-plane-1.0/2.0): zips fail the IMF
v1 file contract ("missing required file: encoder.onnx") — pre-
standard layout. Re-export or mark legacy-incompatible.
3. **ara-diac-small-1.0-int8 output corruption**: spurious 'F' tokens
('...عَلَيْكُمْFمْF') — quantization-era artifact; superseded by the
2.x line anyway.
Loading
Loading