Skip to content

docs: record B3 stage-2 mixed verdict in WO10 - #406

Merged
ronaldtse merged 1 commit into
mainfrom
todo/b3-stage2-verdict
Nov 3, 2026
Merged

ronaldtse merged 1 commit into
mainfrom
todo/b3-stage2-verdict

Conversation

@ronaldtse

Copy link
Copy Markdown
Contributor

Summary

  • run-087-oracle-student completed (5h44m, 3 epochs, loss 0.017); in-job evals: WikiNews-2024 multiref WER 14.0919 (gate < 17.3794 GREEN) but SadeedDiac-25 total DER 7.2471 (gate <= 2.3864 MISSED)
  • Mixed verdict: no export/parity/certify per the all-green stage gate; version untouched
  • Owner options recorded: specialist ship alongside r7, Sadeed-weighted retry, or close-and-bank the durable relabel corpus

Test plan

  • Verdict numbers transcribed from job 6ae7d86f0e0338d1bb7db20c logs
  • No code changes; work-order documentation only

…te missed)

run-087 oracle student: WikiNews-2024 multiref WER 14.0919 vs r7's
17.3794 (gate green, -19%) but SadeedDiac-25 total DER 7.2471 vs the
2.3864 ceiling (gate missed). All-green required: no IMF export, no
parity, no certify; version stays the owner's. Owner options recorded
in the work order.
@ronaldtse
ronaldtse merged commit b2601c5 into main Nov 3, 2026
8 of 12 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant