Why we tried it
Adapt a released tri-view predictor on constrained hardware and test whether gains depend on actions.
What the evidence shows
A 71.06% matched local error reduction still trailed persistence; a separate small prototype improved about 6.54%, while EEF removal performed similarly.
Where the work stands
The historical scored system, adapter and prototype are distinct; ten reused trajectories do not establish causal action contribution or unseen-task validity.
Possible next steps
Use untouched task groups and paired action-divergent branches, vary adapter seeds, and bind exact model and evaluation artifacts.
Suggested next steps are untested research questions, not new experimental results. The archive records manuscript provenance, licensing and status.