Elamur Groupantslangseg-cues
The groupant beside tools/langseg/src/cues.rs in the Mova repository: what the code computes, what it never does, how to go back from its outputs, its invariants, cost and checks.
Facts
Implementation material. A Mova code model describes this code at one source revision: partial models, finite checks and their stated scope. It is not a reviewed general exposition, and its labels are shown as written, never raised.
Components
| component | label | where |
|---|---|---|
| approximant | partial | approximant |
| ast | partial | syntax.json |
| cost | symbolic source derived | invariant |
| descriptant | present | descriptant |
| invariant | partial | invariant |
| tests | executed bounded checks | tests |
Obernants
The typed relations around the forward model: the behaviour complement (what the code never does), the reverse reconstruction (every input compatible with an output) and specialised views.
| id | type | label | where |
|---|---|---|---|
behavior_complement | behavior_complement | partial source derived model | behavior_complement |
reverse_reconstruction | reverse_reconstruction | partial source derived model | reverse_reconstruction |
first_form_wins | first_form_wins | partial source derived model | first_form_wins |
name_alias | name_alias | partial source derived model | name_alias |
affix_threshold | affix_threshold | partial source derived model | affix_threshold |
expectation_weight_separation | expectation_weight_separation | partial source derived model | expectation_weight_separation |
Model text
descriptant
Descriptant
Prospective epoch
2026-10-31 source review predicted: word_cues inserts declaration-order WORDS using entry.or_insert; first duplicate form wins. gloss chooses first matching C:name/A:name or fixed P label. p/s encode affix orientation and minimum length consumed by Featurizer. Resource model: W words,A affixes: map O(W log W) construction/O(W) entries; gloss O(W+A) scan. Static borrowed strings; no filesystem. External prospective.md predates test additions and preserves APIs, effect order, hazards and interaction hypotheses.
Observed epoch
Bounded real helper checks in tests.md constrain these models; syntax is a partial lexical projection with no macro expansion, name resolution or proof. Production expressions unchanged; five inline test modules added. Compiled mutants are temporary and exactly restored. Existing malformed-input and nonfinite domains remain explicit, not silently repaired. Parent composes this scope with siblings without adding overlapping costs twice.
Static representation and bounded downstream checks; all catalog semantics unverified.
approximant
Forward relation
word_cues inserts declaration-order WORDS using entry.or_insert; first duplicate form wins. gloss chooses first matching C:name/A:name or fixed P label. p/s encode affix orientation and minimum length consumed by Featurizer.
Domain: Names may alias spellings. Expectations are explanatory text, not learned weights. Missing labels return None; exact C:/A: prefixes select separate tables.
invariant
Invariant and cost
Invariant
Names may alias spellings. Expectations are explanatory text, not learned weights. Missing labels return None; exact C:/A: prefixes select separate tables.
Cost
W words,A affixes: map O(W log W) construction/O(W) entries; gloss O(W+A) scan. Static borrowed strings; no filesystem.
Limits
Static representation and bounded downstream checks; all catalog semantics unverified. No percentage surrogate is implemented: categorical parsing/effect observations lack an admitted numerical error metric. Any future profile must define its metric/domain and use only1/2/5/10 percent caps.
models
Typed source-bound relations
Subject: exact source SHA-256 in manifest. Domain and resource premises are in invariant.md. Each model is partial, source-derived, over a single helper invocation or the described finite trace. Complements are proposed inner exclusions under those premises; reverse relations retain all compatible antecedents, with no unique inverse claim. The oracle is the named checks in tests.md; unexecuted cases remain hypotheses. No numerical approximation certificate.
behavior_complement
No learning or effect occurs. Missing names cannot fabricate glosses; later duplicate forms cannot replace earlier map entries. Expectations cannot directly choose classes.
reverse_reconstruction
A gloss admits all names whose first matching entry has that text; aliases prevent recovery of spelling. Map forgets later duplicates and their positions.
first_form_wins
map[f]=WORDS[min{i:form_i=f}]. cue-di compiled mutant changes actual feature identity; downstream cue assertion rejects.
name_alias
ืืขื/ืืขื ืช share now_kean; gloss observes first matching name. Returned expectation cannot identify firing surface.
affix_threshold
Affix min_len is checked against Hebrew letters downstream, not byte length. Table itself normalizes nothing.
expectation_weight_separation
Counts determines Model.logp; expect appears in explanations. Changing expect alone cannot change scoring.
tests
Checks and evidence
Named source/helper checks: cue_features_fire_on_pointed_and_unpointed_forms.
Wave31 on2026-10-31:30 distinct exact tests passed twice debug and twice release, with equal sorted named outcomes. This node claims only its listed subset; empty subset means source review only. Six actual source-behavior mutants compiled, failed assertions, restored exact bytes/hash, and passed focused green reruns. No assertion-only mutants; existing swapped-input controls are separate. Receipt bundle: $MOVA_DATA/runs/octopus-wave31-corpus-2026-10-31 (native-commands.json, source-test-map.json, mutation-results.json, .status/.log). No live corpus, teacher, frozen assessment, download or production data used.
Static representation and bounded downstream checks; all catalog semantics unverified.
This page in Markdown โ https://md.groupants.elamur.ai/free/en/mova/tools/langseg/src/cues.rs.groupant/