diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-08-06 16:08:32 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-08-06 16:08:32 -0500 |
| commit | a6d9cc53f77906305dcefa454d77151c1996405f (patch) | |
| tree | 5c1e92c096d2507ce2afa72d755d0f789f157ef6 /PAPER_PLAN.md | |
| parent | 70e545c95f32275d55b132ecd1c5ae167db5e258 (diff) | |
docs: position SDIL as debiasing plug-in
Diffstat (limited to 'PAPER_PLAN.md')
| -rw-r--r-- | PAPER_PLAN.md | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/PAPER_PLAN.md b/PAPER_PLAN.md index 6f40ee0..60ce67e 100644 --- a/PAPER_PLAN.md +++ b/PAPER_PLAN.md @@ -7,7 +7,9 @@ > Propagation, EP and coupled-learning backbones. The manuscript below remains > the evidence-bound account of completed experiments until the new physical > and cross-backbone gates pass; prospective claims must not be copied into it -> as results. +> as results. In this direction SDIL is a debiasing plug-in, so the old Pareto +> figure is not a planned main figure; cost is handled by matched protocols and +> an overhead audit table. This document maps the frozen evidence to a defensible paper narrative. It is not permission to promote development results or to hide failed gates. |