task_name: surface_form_parity
manifest_version: "1.0.0"
issue: 216
task_type: regression-fixture
description: |
  Mixed-provenance regression fixture for the §F.3.6 reviewer-type (surface-form)
  asymmetry, anchored to first-party verbatim examples from Kim et al. 2026
  (arXiv:2605.20668v1). The documented failure is an AI meta-reviewer applying two
  standards keyed off PROSE STYLE: it penalises informal/vague human wording and
  credits technical-precise AI wording. The hook is framing_style, NOT the author
  label — so this targets SURFACE-FORM parity, not authorship parity.

  This is NOT a calibrated detector. There is no deterministic predictor for this
  bias, n is tiny, and the paper's 29-FN-human / 10-FP-AI split is DIRECTIONAL
  (§H), not a calibration target. The validator checks data integrity + provenance
  honesty + pair invariants only — it does not compute FNR/FPR or assert
  distributional asymmetry.

  counterfactual_rewrite items are maintainer-authored minimal-difference variants
  (NOT first-party). They give the prompt discipline "do not let framing style flip
  a correctness verdict" paired, traceable test material.
target:
  entrypoint: scripts.check_surface_form_parity
  gold_set_path: gold_set.json
sample_n: 7
labels: ["informal_vague", "technical_precise"]
provenance_distribution:
  - provenance_type: paper_verbatim
    n: 4
  - provenance_type: counterfactual_rewrite
    n: 2
  - provenance_type: maintainer_boundary
    n: 1
pairs:
  - pair_id: pair-01
    members: ["sfp-001", "sfp-001-cf"]
  - pair_id: pair-03
    members: ["sfp-003", "sfp-003-cf"]
source_paper:
  arxiv_id: "2605.20668v1"
  citation: >
    Kim, S., Yoon, D., Gashteovski, K., Suk, J., Baek, J., et al. (2026).
    On the limits and opportunities of AI reviewers: Reviewing the reviews of
    Nature-family papers with 45 expert scientists. arXiv:2605.20668v1 [cs.CL].
related_issues:
  - issue: 273
    relationship: cross-reference-only
    note: >
      #273 (same-family / rubric-aware calibration) is a DIFFERENT mechanism. It is
      NOT folded into #216 — see docs/design. Cross-reference only; no shared prompt,
      gold, or runtime wiring.