LEARNING / 17 SEP 2026
A trajectory is not a training sample.
Collides
Working thesis
Production trajectories are becoming a commercial continual-learning primitive, but training formats discard information-state, authority, evidential-origin and consequence semantics.
Research questions
- 01
Separate within-trajectory credit from cross-trajectory weighting and outcome truth.
- 02
Require protected-case regression and evaluator-independence checks.
- 03
Preserve authority, consent and evidential origin through every update.
Connected layers
Primary sources