tegwick
6808aa3594
IR-WP-0007-T03 can now record a capture-linked publication separately from generated drafts. The field-outcome design is written, and credential, billing, and website-pull gates were rechecked and left waiting. Assistant: grok Assistant-Session: 01a106a5-fc07-78c2-b64f-8e04cea25daf
1 KiB
Name reward hacking when a taxonomy is written
Nudge. Noted 2026-10-04.
Derived from compendium/04-socher-eureka-machine.md §5, "Name reward hacking in the field-outcome taxonomy."
The operator's Gemini notes call reward hacking the hard problem in autonomous science: an objective harness that measures discovery rather than a model exploit. Pant's review makes the same demand in laboratory language. The checker has to be able to disagree with the system that produced the result.
Outcome records already require independent_verification. IR-WP-0007-T02 already owes a failure taxonomy before any paid sample is called representative. When that taxonomy is written, reward hacking and a verifier that shares the generator's errors are worth a place in it. 2026-10-04: the class is named verifier_dependence in
docs/evidence/2026-10-04-ir-wp-0007-t02-sampling-frame.md. IR-WP-0007-T02
stays wait. This note is still not a task.