Skip to content

EPIC: reliability-placement study — where does contract compliance live, and how far does it transfer? #5

Description

@JeremyJC67

Status

Active roadmap, opened 2026-08-11. Post-v1 follow-on study. This issue is the canonical ordering of
milestones for the next phase.

Question

v1 establishes that observation-contract compliance is measurable and that agents fail it in
characterisable ways. The natural next question is where compliance can be installed, and how far an
installed compliance transfers.

Core claim under test. Reliability is a quantity that can be located. The same contract target can be
installed at different loci in the stack, and the scientific content is where it lives, how much of it
transfers, and at what cost.

LocusMechanismExpected behaviour
promptcontract stated in contextcheapest; degrades under shift and adversarial pressure
scaffoldverifier / repair loop around the agentstrong in-distribution; brittle to unanticipated constraints
weightsprocess-supervised post-training (LoRA-scale)generalises to unseen constraints iff the abstraction was learned, not the form
architecturetyped tool boundary making the violation unrepresentablestrongest guarantee; narrowest coverage

Design: 4 loci x 3 test beds x 3 generalisation distances, over one fixed contract family.
Generalisation distance is the load-bearing factor: (1) same family, unseen instances; (2) same abstract
constraint, disjoint surface form; (3) different abstract constraint.

Milestones

MilestoneIssueDeliverableGate
M0 — reproducibility floor#2CI over the suite and the 33 task manifestscurrent
M1 — perturbation layernot yet filedAny task runnable as (task, perturbation, seed), deterministicallycurrent
M2 — form-disjoint families#4Two families, same constraints, disjoint surface vocabularycurrent
M3 — locus harnessnot yet filedThe same contract target installed at all four loci, one runnerM1, #4
M4 — monitorability metricnot yet filedCounterfactual dependence of outcome on process, per conditionM1
M5 — transfer mapnot yet filedCompliance vs generalisation distance, per locus, with costM0-M4

Milestones are filed as issues when their gate opens, not before.

Dependency order

#2 → (M1, #4) → M3 → M4 → M5

Work outside this chain must state which milestone it unblocks, or carry the deferred label.

Pre-registered predictions

Recorded before any run, so that a null result is reportable:

  • P1 — at distance (1), all four loci look similar. This is why single-locus results in the literature
    read as successes.
  • P2 — at distance (2), the loci separate. This separation is the result; without it there is no
    placement axis.
  • P3 — monitorability predicts weight-locus transfer at distance (2) better than in-distribution
    compliance does.

Falsification

If the loci do not separate at distance (2), the placement claim is withdrawn rather than protected with
post-hoc conditions. If monitorability cannot be measured reliably, P3 is dropped and the study reduces to a
three-factor transfer map.

Exit condition

  • CI green on main and reproducible from a clean checkout
  • perturbation layer determinism test passing
  • two form-disjoint contract families, disjointness check enforced
  • one contract target installed at all four loci in a single runner
  • monitorability measured for every condition, or P3 formally dropped
  • transfer map produced with cost per unit of reliability, and P1/P2 adjudicated in writing

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    epicRoadmap issue: holds a dependency chain and an exit condition

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions