* docs(adr-0084): propose definitional layer + prompt-diversity suite
Three companion artifacts proposing the next substantive design step
after ADR-0083:
1. ADR-0084 (Proposed) — Definitional Layer for Lexicon Packs
Optional `definition` block on pack entries: gloss,
definitional_atoms, predicates_invited, definition_version,
provenance. Pack-level opt-in. Closure rule: every word in a
gloss must resolve to a same-pack lemma, another mounted pack's
lemma, or a primitive in a new `packs/primitives/` pack.
NO composer change in this ADR (sequenced for ADR-0085) —
ratify substrate before any consumer depends on it.
2. evals/prompt_diversity/ (Proposed) — companion eval lane
~50 cases across question-shape × sophistication × domain,
measuring three new metrics: response_shape_fit,
audit_in_surface_rate (quantifies the trust-boundary leak into
user surfaces), gloss_quote_rate (zero today; rises with future
gloss-aware composer). No v1 pass thresholds — the lane
establishes a baseline distribution so future work has
something to move. 26 seed cases authored covering all 21
categories.
3. docs/handoff/ADR-0084-pack-content-brief.md — paste-ready brief
for a cheaper/faster dev agent to produce the pack content in
parallel. Self-contained, 5 sequenced phases (primitives pack
→ extend 9 existing glosses → add to relations/anchors → write
closure verifier → run safety lanes), explicit don't-touch list
(no composer / runtime / algebra / Greek+Hebrew packs / schema
parser), no-LLM-glosses discipline, per-phase acceptance.
Discovery while drafting: 9 packs already carry glosses.jsonl
under language_packs/data/ with a flat schema (78 entries in
en_core_cognition_v1 alone). The brief reflects that — most
work is extending existing entries, not authoring from scratch.
Strategic context: ADR-0083 raised the *depth* ceiling on chain
composition; ADR-0084 raises the *fidelity* ceiling. The φ
separation probe (memory: phi-separation-falsified) established
that semantic capability lives in chain composition, not in φ
geometry, so deepening the composer's substrate is the natural
next step. ADR-0084 → 0085 (gloss-aware composer) → 0086
(predicate licensing at ratification) is the planned sequence.
* feat(adr-0084): substrate — schema parser, primitives loader, closure verifier
Substrate-only code-side for ADR-0084 (Definitional Layer for Lexicon Packs).
No composer touches the new fields yet; consumer integration is ADR-0085.
Schema (additive, default preserves byte-identity)
- LanguagePackManifest.definitional_layer: bool = False
- compiler loader propagates the flag from manifest.json
language_packs/definitions.py (new)
- GlossEntry dataclass: lemma, gloss, pos, definitional_atoms,
predicates_invited, definition_version, provenance_ids
- parse_gloss_entry(payload, *, strict) — strict mode enforces ADR-0084
§Schema validation row-by-row: required keys, typed lists, no
unknown keys, positive definition_version; lax mode preserves the
legacy two-field shape for back-compat
- load_pack_glosses(pack_id, *, strict) with cache + clear hook
- verify_definitional_closure(pack_id, *, mounted_pack_lemmas,
primitive_lemmas, strict) returning tuple[ClosureViolation, ...];
case-insensitive resolution; cycles permitted per ADR
packs/primitives/loader.py (new)
- Sister loader to packs/safety/ and packs/identity/
- PrimitivesPack frozen dataclass with .lemmas frozenset
- Gates: checksum match, kind=='primitives', definitional_layer:true,
never_auto_mutable:true, pack_id matches dir, primitive_count
cross-check, duplicate-lemma rejection, path-traversal rejection,
strict per-entry schema with allow-list
- DEFAULT_PRIMITIVES_PACK = 'en_semantic_primitives_v1'
tests/test_adr_0084_definitional_substrate.py
- 38 tests covering strict parser (each required key rejection, unknown
key rejection, empty predicates_invited allowed, empty
definitional_atoms rejected, invalid definition_version), lax
parser back-compat, load_pack_glosses (missing/strict raise/lax
skip/malformed JSON), closure verifier (same-pack/primitive/mounted/
unresolved/case-insensitive), primitives loader (every gate), and
a back-compat check that every shipped pack still ratifies with
definitional_layer=False
Lanes: smoke 67/0, cognition 120/0/1, teaching 17/0, runtime 19/0,
packs 6/0. Cognition eval byte-identical 100/91.7/100/100.
When the content PR lands (primitives.jsonl + extended glosses.jsonl
under ADR-0084-pack-content-brief.md), the gate catches any closure-rule
violation without further code change.
* feat(evals): prompt_diversity lane runner — measurement instrument for ADR-0084+
Implements the runner against the existing contract.md + 26-case v1
public split. Lane auto-discovered by evals.framework via the standard
contract + runner convention.
Runner (evals/prompt_diversity/runner.py)
- run_lane(cases, *, config, workers) -> LaneReport
- 5 metrics: intent_accuracy, versor_closure_rate (carried over from
cognition), plus the three new lane-specific metrics —
response_shape_fit, audit_in_surface_rate, gloss_quote_rate
- breakdown dict groups by (question_shape, sophistication, domain)
per contract §How to read the output
- mirrors evals.cognition.runner's parallel worker pattern
Per-shape classifier (deliberately substring/regex-simple at v1)
- predicate_identity, explanation, sequence, two_subject_contrast,
narrative, honest_disclosure
- Unknown shape => neutral pass (don't penalise new categories)
Audit-leak detector
- trust-boundary preamble markers (teaching-grounded (, pack-grounded
(, No session evidence yet.)
- dotted semantic-domain tag regex (cognition.illumination, etc.)
Gloss-quote detector
- resolves expected_terms via chat.pack_resolver.resolve_gloss
- 4-token contiguous-window match against surface (high-confidence
"gloss actually quoted", not "shared one common word")
Tests (tests/test_prompt_diversity_runner.py — 23)
- shape classifier parametrized over the six expected_shape values
- audit-leak detector parametrized over preamble + tag + clean cases
- end-to-end on v1 public:
* versor_closure_rate == 1.0 (only v1 pass threshold per contract)
* every metric in [0, 1]
* breakdown groups present with the four per-cell metrics
* diversity gate: >=5 question shapes, >=3 domains
(defends against future regressions that collapse the suite
back to a cognition-shaped fixture)
v1/public baseline (26 cases)
intent_accuracy : 65.4% (contract predicted 70-85%)
versor_closure_rate : 100.0% (only v1 pass threshold) PASS
response_shape_fit : 53.8% (contract predicted low)
audit_in_surface_rate: 42.3% (contract predicted ~100%)
gloss_quote_rate : 7.7% (contract predicted 0%)
Three baseline surprises worth noting in the report (NOT failures —
the v1 lane is explicitly there to establish the distribution):
- audit_in_surface_rate at 42% (not 100%) means the chain-walk leak
fires on ~11/26; the other 15 are honest-disclosure cases that
emit no audit envelope. Sharpens the future surface-vs-envelope
ADR's actual target: grounded surfaces specifically.
- response_shape_fit at 54% (not "low") — classifier likely has
false positives on the ", which " cause-marker. Worth tightening
once we have an ADR-0085 baseline to compare against.
- intent_accuracy at 65% (below predicted 70-85%) — classifier dips
harder on adversarial/cross-pack than expected. Real gap.
All five smoke/cognition/teaching/runtime/packs lanes still green;
core eval cognition byte-identical 100/91.7/100/100.
* feat(packs): ADR-0084 pack content (primitives + extend glosses + closure verifier) (#65)
* feat(packs): ADR-0084 pack content
* feat(packs): repair ADR-0084 definitional content
* test(adr-0084): adjust substrate manifest tests for post-#65 content reality
PR #65 flipped definitional_layer:true on 13 English packs (9 core +
4 relations + collapse-anchors). The substrate's previous test
test_existing_packs_unchanged asserted that en_core_cognition_v1 +
en_core_relations_v1 still had definitional_layer:False — which was
the right pre-content invariant but is wrong post-content.
Replace it with two complementary tests that hold against real content:
- test_non_opted_packs_default_false:
pins that packs that DIDN'T flip the flag (en_minimal_v1,
he_core_cognition_v1, grc_logos_cognition_v1) still surface
definitional_layer=False through the loader. Defends against
a future change accidentally flipping the flag on a non-opted
pack.
- test_opted_packs_carry_flag:
pins that packs that DID flip the flag (en_core_cognition_v1,
en_core_relations_v1) surface definitional_layer=True through
the loader. Proves the substrate's manifest-field propagation
works against real ratified content, not just fixture packs.
Net: +1 test, same intent (substrate ratifies the manifest field
correctly), now with real-content coverage on both sides of the gate.
All 62 ADR-0084 substrate + prompt-diversity tests pass.
196 lines
6.9 KiB
Python
196 lines
6.9 KiB
Python
"""
|
|
Schemas for CORE compiled linguistic manifolds.
|
|
|
|
A language pack is not a dataset. It is a deterministic, checksummed,
|
|
compiled linguistic manifold: lexical surfaces, morphology, grammar
|
|
attractors, cross-language resonances, and holonomy-level proof cases.
|
|
|
|
These schemas intentionally do not load corpora. They define the contract the
|
|
Supervised Seeding Epoch must satisfy before Hebrew and Koine Greek gates can
|
|
engage.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from dataclasses import dataclass, field
|
|
from enum import Enum
|
|
from typing import Mapping, Sequence
|
|
|
|
|
|
class LanguageRole(str, Enum):
|
|
"""Architectural role of a language pack in CORE-Logos."""
|
|
|
|
OPERATIONAL_BASE = "operational_base"
|
|
ARTICULATION_SURFACE = "articulation_surface"
|
|
DEPTH_ROOT = "depth_root"
|
|
DEPTH_RELATION = "depth_relation"
|
|
|
|
|
|
class OOVPolicy(str, Enum):
|
|
"""Out-of-vocabulary behavior for a pack."""
|
|
|
|
FAIL_CLOSED = "fail_closed"
|
|
TAGGED_FALLBACK = "tagged_fallback"
|
|
PROPOSE_VOCAB_EXPANSION = "propose_vocab_expansion"
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class LanguagePackManifest:
|
|
"""Pinned manifest for one compiled language pack."""
|
|
|
|
pack_id: str
|
|
language: str
|
|
role: LanguageRole
|
|
script: str
|
|
normalization_policy: str
|
|
source_manifest: str
|
|
determinism_class: str
|
|
checksum: str
|
|
version: str = "1.0.0"
|
|
gate_engaged: bool = False
|
|
oov_policy: OOVPolicy = OOVPolicy.FAIL_CLOSED
|
|
# Optional dual-checksum for the companion ``glosses.jsonl`` file.
|
|
# When present, the loader verifies the bytes-on-disk match this
|
|
# SHA-256 just like the lexicon checksum. Absent on legacy packs
|
|
# that ship no glosses (back-compat — never raised in that case).
|
|
# Glosses are an additive overlay; bumping ``glosses_checksum`` does
|
|
# NOT perturb the immutable ``checksum`` (lexicon seal).
|
|
glosses_checksum: str | None = None
|
|
# ADR-0084 — pack-level opt-in for the definitional layer. When
|
|
# True, every gloss entry must carry the extended schema
|
|
# (``definitional_atoms``, ``predicates_invited``,
|
|
# ``definition_version``) and pass the closure rule. Default False
|
|
# leaves every existing pack byte-identical.
|
|
definitional_layer: bool = False
|
|
|
|
def __post_init__(self) -> None:
|
|
if not self.pack_id:
|
|
raise ValueError("LanguagePackManifest.pack_id is required.")
|
|
if not self.language:
|
|
raise ValueError("LanguagePackManifest.language is required.")
|
|
if not self.checksum:
|
|
raise ValueError("LanguagePackManifest.checksum is required.")
|
|
if self.role in {LanguageRole.DEPTH_ROOT, LanguageRole.DEPTH_RELATION}:
|
|
if self.gate_engaged and self.oov_policy is not OOVPolicy.FAIL_CLOSED:
|
|
raise ValueError(
|
|
"Depth packs must fail closed while gate_engaged=True; "
|
|
"unknown Hebrew/Greek surfaces must not collapse to a fallback point."
|
|
)
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class MorphologyEntry:
|
|
"""
|
|
Morphological decomposition for a surface form.
|
|
|
|
Ordering is load-bearing. For Semitic root morphology and Koine grammar,
|
|
non-commutative composition means prefix/stem/inflection/suffix order must
|
|
be preserved exactly.
|
|
"""
|
|
|
|
morphology_id: str
|
|
surface: str
|
|
lemma: str
|
|
language: str
|
|
root: str | None = None
|
|
prefix_chain: tuple[str, ...] = field(default_factory=tuple)
|
|
stem: str | None = None
|
|
inflection: Mapping[str, str] = field(default_factory=dict)
|
|
suffix_chain: tuple[str, ...] = field(default_factory=tuple)
|
|
|
|
def __post_init__(self) -> None:
|
|
if not self.morphology_id:
|
|
raise ValueError("MorphologyEntry.morphology_id is required.")
|
|
if not self.surface:
|
|
raise ValueError("MorphologyEntry.surface is required.")
|
|
if not self.lemma:
|
|
raise ValueError("MorphologyEntry.lemma is required.")
|
|
if not self.language:
|
|
raise ValueError("MorphologyEntry.language is required.")
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class LexicalEntry:
|
|
"""One surface/lemma entry in a compiled linguistic manifold.
|
|
|
|
`epistemic_status` follows ADR-0021: it is a *position in the
|
|
revision graph*, not a source-trust tier. The default is
|
|
``"speculative"`` per ADR-0021 §Schema impact: "transitions to
|
|
COHERENT / CONTESTED / FALSIFIED only via the review path." A pack
|
|
lexicon row that wants to be admissible as evidence
|
|
(``ADMISSIBLE_AS_EVIDENCE``) must declare
|
|
``"epistemic_status": "coherent"`` explicitly; the declaration is
|
|
itself the curator's stamp. Pack authority alone is not coherence
|
|
judgment — defaulting unmarked rows to COHERENT would re-import the
|
|
bias ADR-0021 refuses (see ``docs/truth_seeking_schema.md`` §1).
|
|
"""
|
|
|
|
entry_id: str
|
|
surface: str
|
|
lemma: str
|
|
language: str
|
|
part_of_speech: str | None = None
|
|
pos: str | None = None
|
|
morphology_id: str | None = None
|
|
morphology_tags: tuple[str, ...] = field(default_factory=tuple)
|
|
semantic_domains: tuple[str, ...] = field(default_factory=tuple)
|
|
manifold_point_checksum: str | None = None
|
|
provenance_ids: tuple[str, ...] = field(default_factory=tuple)
|
|
epistemic_status: str = "speculative"
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class GrammarAttractor:
|
|
"""
|
|
Structural grammar attractor seeded into the shared manifold.
|
|
|
|
Morphology is operator composition. Semantic domain is attractor geometry.
|
|
Alignment is resonance. This class represents the attractor layer only.
|
|
"""
|
|
|
|
attractor_id: str
|
|
language: str
|
|
role: str
|
|
description: str
|
|
operator_order: tuple[str, ...] = field(default_factory=tuple)
|
|
checksum: str | None = None
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class AlignmentEdge:
|
|
"""Weighted directional resonance between entries or concepts."""
|
|
|
|
source_id: str
|
|
target_id: str
|
|
relation: str
|
|
weight: float
|
|
evidence_ids: tuple[str, ...] = field(default_factory=tuple)
|
|
|
|
def __post_init__(self) -> None:
|
|
if not 0.0 <= self.weight <= 1.0:
|
|
raise ValueError("AlignmentEdge.weight must be in [0, 1].")
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class HolonomyAlignmentCase:
|
|
"""
|
|
Crown proof case for the three-language design.
|
|
|
|
The language system succeeds when aligned canonical clauses produce nearby
|
|
holonomies without flattening their distinctions. This is not token-level
|
|
translation; it is dynamic field-path resonance.
|
|
"""
|
|
|
|
case_id: str
|
|
description: str
|
|
source_refs: tuple[str, ...]
|
|
pack_ids: tuple[str, ...]
|
|
expected_relation: str
|
|
negative_source_refs: tuple[str, ...] = field(default_factory=tuple)
|
|
tolerance: float | None = None
|
|
|
|
def __post_init__(self) -> None:
|
|
if len(self.source_refs) < 2:
|
|
raise ValueError("HolonomyAlignmentCase requires at least two source_refs.")
|
|
if len(self.pack_ids) < 2:
|
|
raise ValueError("HolonomyAlignmentCase requires at least two pack_ids.")
|