coulomb/frontend-patterns
5
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Incorporate submitted pilot review and adjudicate four score differences

Browse source 
Assistant: codex
Assistant-Model: gpt-6-astra
Assistant-Session: 01a0cffe-206f-7323-89d3-b20ca4cf4bf9
This commit is contained in:
tegwick 2026-09-26 21:06:38 +02:00
parent 629bc2e880
commit 21030784b1
13 changed files with 659 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
research/plan/OpenQuestions.md
View file

@ -65,3 +65,9 @@ under two equivalent-information conditions. Q04 remains open: small mixed descr
differences, same-author editorial scoring and pending human adjudication do not
establish usefulness. The six briefs are consumed pilot material, not future held-out
items. Next: independent ratings before changing guidance or replicating.
2026-09-26 submitted-review update: [adjudicated totals](../evaluations/language-pilot-01/submitted-review-01/adjudication.md)
are 18/24 for each condition after four corrections. No presentation benefit is
established; equal totals do not establish equivalence. Worksheet declares model
authorship and prior aggregate-result exposure. Q04 remains open; future rubrics
should split compound obligations before execution.