Elamur Groupantsmath-solver-orchestration-memory-fallback
The groupant beside math/src/solve.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 |
| composition | partial | composition |
| descriptant | present | descriptant |
| invariant | partial | invariant |
| tests | partial bounded evidence | 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 |
failure_precedence | failure_precedence | partial source derived model | failure_precedence |
representation_transport | representation_transport | partial source derived model | representation_transport |
resource_boundary | resource_boundary | partial source derived model | resource_boundary |
domain_boundary | domain_boundary | partial source derived model | domain_boundary |
Model text
descriptant
Source-bound description
Memos filter verified B then join Problem by id. C trains10epochs, filters same method/count/template/not own id, ranks skeleton and tries<=3. Multiple-choice requires unique computed numeric option. Verified candidate wins, else copy selected B fields with by_slm false.
Epoch 2026-10-31: prospective model before bounded synthetic tests; production behavior unchanged. Source scope bytes[9085,13850). Teacher/Lean loops and append-failure path unexecuted; no accuracy claim. Not every branch in this fragment executed.
approximant
Forward relation
Memos filter verified B then join Problem by id. C trains10epochs, filters same method/count/template/not own id, ranks skeleton and tries<=3. Multiple-choice requires unique computed numeric option. Verified candidate wins, else copy selected B fields with by_slm false.
Domain: Synthetic bounded records and templates; finite similarity/margins. Equal similarity preserves memo order.
Executable Rust approximant unsupported: categorical/effect observations have no useful percentage metric yet. Future caps1/2/5/10% only; no automatic substitution.
invariant
Invariant and cost
Missing Problem id excludes memo. Empty memory excludes by_slm success. No held split filter here; caller must enforce.
Ordering: Config/verified→id lookup→template; C margin→candidates→instantiate→option selection→gate→fallback.
Cost: Perceptron train+per-problem memory feature/skeleton work and<=3 plan runs (2 each with options). O(candidate list+records+weights).
composition
Composition and epoch memory
Correct flag copied for reporting but does not control memo admission; internal gate distinct from grade.
Source scope bytes[9085,13850). Teacher/Lean loops and append-failure path unexecuted; no accuracy claim. Not every branch in this fragment executed.
Partial syntax and finite tests do not establish all caller assumptions.
models
behavior_complement
Missing Problem id excludes memo. Empty memory excludes by_slm success. No held split filter here; caller must enforce.
reverse_reconstruction
Memo can derive from verified incorrect record. C fallback cannot reconstruct omitted history/trace/error; template success records donor id but not entire training sequence.
failure_precedence
Config/verified→id lookup→template; C margin→candidates→instantiate→option selection→gate→fallback.
representation_transport
Correct flag copied for reporting but does not control memo admission; internal gate distinct from grade.
resource_boundary
Perceptron train+per-problem memory feature/skeleton work and<=3 plan runs (2 each with options). O(candidate list+records+weights).
domain_boundary
Synthetic bounded records and templates; finite similarity/margins. Equal similarity preserves memo order.
tests
Evidence, 2026-10-31
Selected actual-source checks: cargo test -p math --lib solve::wave28. Exact commands, numeric statuses, normalized outcomes, compiled assertion controls and restoration receipts are in the external wave28 mathcontrol run.
Source scope bytes[9085,13850). Teacher/Lean loops and append-failure path unexecuted; no accuracy claim. Not every branch in this fragment executed.
No live teacher/API, trained model, corpus/held-out, GPU, DB or expensive benchmark run. Parent syntax generated twice identically. Scope remains partial.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/math/src/solve.rs.groupant/memory-fallback.groupant/