Skip to content

OATH verifier: close the trigger-recall gap (52% of decimals sit on unbound lines) #39

Description

@fathomlab

The problem: our claim verifier (styxx.certify, the OATH implementation) only obligates a number to be grounded when its line contains recognized trigger vocabulary (AUROC, margin, rate, accuracy...). Numbers on lines without those words land in ABSTAIN - honestly reported, but unchecked.

We measured the size of this gap: 0.5227 of full-precision decimals in our own corpus sit on unbound lines (b35-c G2b). That is the largest known hole in the verifier.

Why it is hard (and interesting): naive fixes over-trigger. Our own history is instructive - a blunt trigger add produced 6 false accusations (cycle 23), and a bare value-range guard let an ordinal collide with a correlation boundary (cycle 24). The v0.4 decimal+range guard was the shipped compromise. See the comments in styxx/certify.py around _TRIGGERS and _TRIGGERS_CORR.

What a contribution looks like:

  1. A proposal for broader obligation (dependency parse? sentence-level classifier? proximity to a receipt-name mention?)
  2. Run it against the existing mutant battery: python papers/closed-model-frontier/validate_oath_v0.py
  3. Report catch rate and false-accusation rate - a fix that raises catches while adding a single false accusation does not ship (that rule is why the verifier is trustworthy)

The rule of this repo: any change to certify.py needs its own preregistration with frozen bars before the battery runs. See PREREG_oath_v062_signed_extraction for the shape.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or requesthelp wantedExtra attention is needed

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions