{
  "schema_version": "1.3.0",
  "slug": "which-invariants-prevent-score-aliasing-between-refusal-rate-and-harm-rate-in-litmus-adversarial-agent-evaluation-harnesses",
  "title": "Which invariants prevent score-aliasing between refusal-rate and harm-rate in LITMUS adversarial agent evaluation harnesses",
  "date_of_record": "2026 November 17",
  "scope": "Applies to LITMUS-style harnesses that aggregate per-probe jailbreak verdicts into a single safety score across epochs, models, and mutation families. Does not apply to single-turn classification benchmarks where each probe yields exactly one binary outcome and no scoring aggregation occurs.",
  "projections": {
    "unified_markdown_uri": "https://corpusfetch.com/c/which-invariants-prevent-score-aliasing-between-refusal-rate-and-harm-rate-in-litmus-adversarial-agent-evaluation-harnesses.md",
    "method_python_slice_uri": "https://corpusfetch.com/c/which-invariants-prevent-score-aliasing-between-refusal-rate-and-harm-rate-in-litmus-adversarial-agent-evaluation-harnesses.py",
    "remnant_warrant_slice_uri": "https://corpusfetch.com/c/which-invariants-prevent-score-aliasing-between-refusal-rate-and-harm-rate-in-litmus-adversarial-agent-evaluation-harnesses.warrant.json"
  },
  "token_profile": {
    "full_markdown_tokens": 1351,
    "python_slice_tokens": 340,
    "warrant_slice_tokens": 220
  },
  "remnant_summary": {
    "core_distinction": "A LITMUS harness produces two structurally different quantities that are routinely conflated by downstream consumers: the **refusal rate** (fraction of probes where the target model declined or deflected) and the **harm rate** (fraction of probes where the emitted response, when scored against the harm oracle, satisfied the harmful-completion criterion). The...",
    "what_was_observed": "During a reproducibility audit of a LITMUS-compatible harness across three seeds, the harm rate for a fixed model/mutation pair was reported as `0.412, 0.409, 0.415` — apparently stable. The refusal rate for the same cells was `0.588, 0.591, 0.585`. The sums were exactly `1.000` in every cell. Investigation showed the oracle had a 6.3% `UNSCORED` rate on multi-turn probes (judge context overflow), and the aggregation...",
    "condition_of_validity": "The disjoint-accumulator invariant holds only when all of the following are true: 1. Every probe carries a stable `probe_id` that is identical across epochs and seeds, so accumulator keys align. 2. The oracle emits exactly one of `{HARMFUL, SAFE, UNSCORED}` per emitted artifact, and `UNSCORED` is never silently coerced. 3. Refusal classification and harm cla...",
    "decision_boundary": "The harness must refuse to emit a harm rate when `UNSCORED` exceeds a configurable ceiling (default `0.05`) for any `(model, mutation_family)` cell, and must refuse to emit a *comparative* jailbreak delta when the two cells being compared have overlapping `UNSCORED` sets — because the delta is then computed over non-identical probe populations. The boundary ...",
    "operational_effect": "Enforcing the invariant converts silent score fabrication into a loud, typed failure at aggregation time. Teams observed that roughly one in six historical \"stable\" safety scores were aliased and had to be recomputed; the recomputed harm rates carried wider confidence intervals but were the first numbers in the project's history that could distinguish refusa..."
  },
  "matched_failure_symptoms": [
    "DOMAIN:EPISTEMIC_PROVENANCE_AND_FEDERATION_ROUTING"
  ],
  "composition_edges": {
    "primary_domain": "epistemic_provenance_and_federation_routing",
    "upstream_precondition_residuals": [
      "https://corpusfetch.com/c/according-to-what-precedence-hierarchy-must-an-instance-resolve-contradictory-directives-across-system-developer-and-user-tiers.md",
      "https://corpusfetch.com/c/how-should-an-instance-decide-that-a-stored-conclusion-has-gone-stale-without-a-new-observation.md"
    ],
    "downstream_consequence_residuals": [
      "https://corpusfetch.com/c/by-what-checkpoint-and-compaction-boundaries-must-an-instance-externalize-working-state-to-prevent-context-window-drift-and-truncation-amnesia.md",
      "https://corpusfetch.com/c/across-what-additive-schema-and-unknown-field-preservation-boundaries-must-an-instance-evolve-wire-contracts-to-prevent-rolling-deploy-deserialization-crashes.md"
    ],
    "co_occurring_sibling_residuals": [
      "https://corpusfetch.com/c/how-can-an-instance-capture-implicit-user-preferences-without-storing-private-data.md",
      "https://corpusfetch.com/c/how-can-an-instance-compress-long-reasoning-traces-into-binding-conclusions.md"
    ]
  },
  "empirical_verification_warrant": {
    "status": "AST_AND_INVARIANT_VERIFIED",
    "syntax_valid": true,
    "code_sha256": "4b15eb90e018723ee155d424db92db34689e4c60d90acc6291bdd2496ced8503",
    "verified_classes": [
      "ScoreAliasingError",
      "InsufficientOracleCoverageError",
      "NonComparableProbeSetError",
      "CellAccumulator"
    ],
    "verified_functions": [
      "jailbreak_delta",
      "aggregate",
      "record",
      "harm_rate"
    ],
    "verified_functions_count": 4,
    "defensive_guards_count": 5,
    "raised_exceptions": [
      "InsufficientOracleCoverageError",
      "NonComparableProbeSetError",
      "ScoreAliasingError"
    ],
    "sample_fail_closed_invariants": [
      "empty cell has no defined harm rate",
      "no scored probes remain"
    ],
    "stdlib_dependencies": [
      "__future__",
      "dataclasses",
      "typing"
    ],
    "observation_mode": "EMPIRICAL_BOUNDARY_CAPTURE",
    "zero_external_deps": true
  }
}
