relower: map and docstring stop reading Accepted/Regressed as findings while the floor is uncalibrated
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
This commit is contained in:
parent
55638d5955
commit
d7ca223bd8
2 changed files with 5 additions and 5 deletions
|
|
@ -1634,8 +1634,8 @@ impl RunStores {
|
|||
|
||||
/// A diagnostic like [`recall`](Self::recall) and
|
||||
/// [`stance_ablation`](Self::stance_ablation), never a gate: it serves the
|
||||
/// distil decision, whether a tuned lowerer beats or matches the prompted
|
||||
/// one. Asks `arm` the byte-identical lowering prompts of round `n`'s live
|
||||
/// distil decision, whether a tuned lowerer is worth keeping against the
|
||||
/// prompted one (calibration-pending: see the map). Asks `arm` the byte-identical lowering prompts of round `n`'s live
|
||||
/// Validation items, reads the arm and the round's own lowering
|
||||
/// (`recorded`) against the items, and, given a `floor` arm (a second
|
||||
/// prompted lowering, so the floor is lowering-only noise), decides by
|
||||
|
|
|
|||
|
|
@ -371,9 +371,9 @@ own Validation reading, which scores those items 0. An arm answer that is not a
|
|||
usable lowering scores 0 and counts in `unusable=`; the same for the floor arm,
|
||||
counted after `floor ... unusable=`. The floor is the spread of the recorded
|
||||
lowering and a second prompted arm (lowering-only noise, no encoder noise). The
|
||||
verdict is `decide_panel` over (recorded, arm, floor): Accepted means the arm
|
||||
helped, Rejected(Regressed) means it is worse, Rejected(NoGain) means it did
|
||||
not clearly help. The rule is calibration-pending: a one-pair floor is noisy
|
||||
verdict is `decide_panel` over (recorded, arm, floor): Accepted is the rule's
|
||||
name for a gain past the floor, Rejected(Regressed) for a loss past it,
|
||||
Rejected(NoGain) for neither. None is yet a finding about the arm. The rule is calibration-pending: a one-pair floor is noisy
|
||||
(a simulated prompted repeat reads NoGain only about 7-15% of the time, Regressed
|
||||
44-60%, Accepted 33-41%), so NoGain is not yet evidence of non-inferiority and
|
||||
is not yet the bar for cut 13; the calibration arms (a prompted repeat, a
|
||||
|
|
|
|||
Loading…
Reference in a new issue