---
title: "world-rerank — Mova code model"
description: "Mova code model of world/src/rerank.rs: components with their labels, obernants and the model text."
url: "https://groupants.elamur.ai/free/en/mova/world/src/rerank.rs.groupant/"
---
Elamur: [Mova](https://md.mova.elamur.ai/free/en/) · [Dictionary](https://md.dictionary.elamur.ai/free/en/) · [Research](https://md.research.elamur.ai/free/en/) · **Groupants**

This page as HTML: <https://groupants.elamur.ai/free/en/mova/world/src/rerank.rs.groupant/>

# world-rerank

The groupant beside world/src/rerank.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

- **kind:** groupant
- **status:** partial source bound model
- **source:** [`world/src/rerank.rs`](https://github.com/ElamurAI/mova-free/blob/main/world/src/rerank.rs)
- **source SHA-256:** `7734d1b33c02f022…`
- **repository folder:** [`world/src/rerank.rs.groupant`](https://github.com/ElamurAI/mova-free/tree/main/world/src/rerank.rs.groupant)
- **repository commit:** `855da7bd4a7b`

> **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 source model | [forward](#forward) |
| **ast** | partial | `syntax.json` |
| **composition** | partial source model | [composition](#composition) |
| **cost** | partial source model | [cost](#cost) |
| **descriptant** | partial source model | [epoch-memory](#epoch-memory) |
| **invariant** | partial source model | [invariant](#invariant) |
| **tests** | partial source model | [checks](#checks) |

## 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 |
|---|---|---|---|
| `world-rerank-behavior-complement` | behavior_complement | partial source model | [behavior-complement](#behavior-complement) |
| `world-rerank-reverse-reconstruction` | reverse_reconstruction | partial source model | [reverse-reconstruction](#reverse-reconstruction) |
| `world-rerank-coupling` | constraint_relation | partial source model | [coupling](#coupling) |
| `world-rerank-observation-fiber` | information_loss | partial source model | [observation-fiber](#observation-fiber) |

## Parts

- [world-rerank-selection](https://md.groupants.elamur.ai/free/en/mova/world/src/rerank.rs.groupant/selection.groupant/)
- [world-rerank-measurement](https://md.groupants.elamur.ai/free/en/mova/world/src/rerank.rs.groupant/measurement.groupant/)
- [world-rerank-repair-boundary](https://md.groupants.elamur.ai/free/en/mova/world/src/rerank.rs.groupant/repair-boundary.groupant/)

## Model text


### model

## world-rerank

### forward

Candidate reranking and token-weighted paired measurement with separate lexical/role repair entrypoints. Pure selection and metric helpers are exercised on supplied candidates; parser/gold I/O and repair inference remain external.

### invariant

Only declared bounded valid domains have modeled invariants. Successful allocations and stable dependency snapshots are assumed; arbitrary public states and external effects are not certified.

### cost

C candidates and N sentences: pick O(C), report O(sum C + 1000N) time and O(N) bootstrap storage. Lambda sweep multiplies candidate cost by ten. repairs repeatedly scans nodes and matrix lookups; hand passes bounded by two. Parser, matrix initialization and induced-rule costs not bounded here. No timing gain claimed.

### composition

main.rs absurd-rerank calls run, which loads dev/test via items using annotator.annotate/annotate_kbest and absurd::judge. Scores pass through en::annotate::annotate_kbest from parser.parse_kbest (f32 promoted to f64), so negative infinity is admissible. run selects lambda on dev, report uses get for sentinel safety. Actual repair callers are tree/reader and evaluation paths; source inspection does not execute them.

### epoch-memory

2026-10-30 first source-bound epoch. Before repair, native invalid-score and cardinality assertions failed on old production behavior. Narrow input checks now prevent undefined NaN ordering and silent report undercount. Infinite parser scores remain supported. Empty conventions and malformed-metadata/repair limits are retained explicitly. No optimization or percentage-error surrogate is proposed.

### checks

cargo test -p world --lib rerank::wave10_models -- --test-threads=1. See root evidence.md for raw evidence hashes, repeated profiles and compiled controls. Syntax is partial; macro bodies are opaque and semantic AST/name/type resolution is absent.

### behavior-complement

Subject/revision: bound source hash and this scope. Domain/observation/operator: Universe: bounded supplied candidate or Row inputs and one terminating call. Exclude observations forbidden by the scoped child invariants, conditional on each child's explicit premises. This is an inner absent-behavior model, not the complement of all Rust traces.

Status: partial source-derived relation; finite evidence only as named in checks. Assumptions: bounded counts without integer or floating arithmetic overflow, successful allocations and fixed dependencies. Unknown external behavior is not excluded. Reverse relations retain ambiguity; finite witnesses do not establish universal completeness.

### reverse-reconstruction

Subject/revision: bound source hash and this scope. Domain/observation/operator: Keep every compatible scoped input and intermediate observation satisfying forward constraints. Aggregate scores, selected candidates and sampled records discard information; no unique inverse or causal explanation follows.

Status: partial source-derived relation; finite evidence only as named in checks. Assumptions: bounded counts without integer or floating arithmetic overflow, successful allocations and fixed dependencies. Unknown external behavior is not excluded. Reverse relations retain ambiguity; finite witnesses do not establish universal completeness.

### coupling

Subject/revision: bound source hash and this scope. Domain/observation/operator: main.rs absurd-rerank calls run, which loads dev/test via items using annotator.annotate/annotate_kbest and absurd::judge. Scores pass through en::annotate::annotate_kbest from parser.parse_kbest (f32 promoted to f64), so negative infinity is admissible. run selects lambda on dev, report uses get for sentinel safety. Actual repair callers are tree/reader and evaluation paths; source inspection does not execute them.

Status: partial source-derived relation; finite evidence only as named in checks. Assumptions: bounded counts without integer or floating arithmetic overflow, successful allocations and fixed dependencies. Unknown external behavior is not excluded. Reverse relations retain ambiguity; finite witnesses do not establish universal completeness.

### observation-fiber

Subject/revision: bound source hash and this scope. Domain/observation/operator: Native fixtures and caller inspection are separate evidence. A checked leaf does not prove parser, filesystem, gold alignment or answer-generation joins. The scope deliberately excludes full assessment and performance claims.

Status: partial source-derived relation; finite evidence only as named in checks. Assumptions: bounded counts without integer or floating arithmetic overflow, successful allocations and fixed dependencies. Unknown external behavior is not excluded. Reverse relations retain ambiguity; finite witnesses do not establish universal completeness.

### evidence

#### Evidence

2026-10-30, synthetic native inputs only. Four rerank and four rreport tests pass twice per debug/release profile; normalized identities/results match. Old production code failed two regression assertions (6 passed, 2 failed); no parser/gold data were loaded. Selection control reverses the primary comparison; aggregation control subtracts the merge score. Both compiled and failed their named independent-oracle assertions; exact source bytes restored and all eight checks passed again.

Bounds: 2916 candidate selections, 216 aggregation fixtures, nine aggregate-fiber pairs, 23 sampling bounds and nine bootstrap sizes plus translation checks. No measured speed or percentage approximation claim.

Raw logs: $MOVA_DATA/runs/octopus-wave10-observation-2026-10-30/. Hashes below bind the recorded bytes; sidecar source hashes bind production plus tests.

- before.log: `a36df3771f5026a26611ceab683f8963d5be83d04a098fe88b76ba656cc486f0`
- debug1.log: `03c1c4c9b65c9febb88ca7af82370b4c0680c14eeec0254db205b63408da5767`
- debug2.log: `a75666665c9ca24cb96d60905077a60ee8b0bed55e764be194b332248abd2716`
- release1.log: `1c7402f41297ed9e7abc53dc6d51fa1ce379aa18c3ae4b7d0d8522e027550a22`
- release2.log: `ebe4c6c3dc5ebb9f6a8b1542efdd6dce3b4f593dda6b48c6fccc68e244cd7d0f`
- debug-normalized.txt: `7caab49871ee32d5168e0f198a4808e623248f70c3f814b41512dd4529cf8410`
- release-normalized.txt: `7caab49871ee32d5168e0f198a4808e623248f70c3f814b41512dd4529cf8410`
- mutant-rerank.log: `1ce1e752636ad2b450c178808c21083daaf4d8249dc2789e94b640fca51f5e01`
- mutant-rreport.log: `70fe2d027578c2226dba324e1ffa494564e7ce453fa9bd978e833a497dd9244f`
- restored.log: `87b391d3a572a7f84f690a8df80bcadd6349e7f5f2dad4bbfee5be656e5fe97f`

Structural validation: both source syntax projections generated twice with identical bytes. Full original registry plus an external scaffold and the two new entries validates twice identically (200 file sidecars,179 fragments, no diagnostics); changed-source gate reports both sources and no missing entries. The scaffold parent is deliberately unmaterialized; central parent integration is separate. The earlier two-entry-only registry failed on unrelated existing sidecar ownership and was replaced by the full registry scaffold. No central registry or parent was edited here. git diff --check passes.