# Goal 5 summary: lossless macro and motif compression

**Integration status:** `complete`

## Scientific boundaries

- Pure EML-DAG and both e-graph-selected EML-DAG tracks contain only pure EML nodes.
- Macro and motif nodes are compression vocabulary nodes, not pure EML nodes.
- Macro graphs are structurally close to labeled compiler/AST graphs and must not be described as single-operator EML.
- Safe-real nontrivial cohorts use branch-insensitive finite-real rewrites; domain-aware cohorts are conditional positive-real-formal results under each row's recorded assumptions.
- Standalone graph MDL and dictionary-inclusive motif MDL are distinct scopes and are never combined into one aggregate.
- Neural-ranker speedup is scoped only to candidate cost scoring; it is not an end-to-end pipeline or mathematical-reasoning speedup.
- Every failure, unsupported case, validation failure, and missing resource observation remains visible in its all-attempted denominator.
- All-processed denominators remain track-specific: the Goal 4 e-graph study covers its frozen 30,000-expression selection, while issue 5-8 covers the full 250,000-expression corpus. Exact cross-track comparisons use the separately joined nontrivial cohorts.

## Frozen Goal 6 export

| Contract | Frozen value |
|---|---|
| Export root | `outputs/final/goal5/export` |
| Run directory | `outputs/final/goal5/export/run-{run_digest}` |
| Completion file | `run.complete.json` |
| Production manifest schema | `geml-goal5-production-export-v1` |
| Model feature schema | `geml-goal5-model-features-v1` |
| Hierarchy schema | `geml-goal5-hierarchy-v1` |

## Production export audit

| Batches | Expressions | Graphs | Hierarchies | Validation failures | Reconstruction failures |
|---:|---:|---:|---:|---:|---:|
| 2500 | 250000 | 1250000 | 250000 | 0 | 0 |

- Subset labels: unavailable — issue 5-8 freezes explicit-only-default-empty subset labels; Goal 4 nontrivial membership is joined externally by exact expression ID
- Runtime: unavailable — the issue 5-8 production completion and batch schemas do not publish per-graph or run runtime observations
- Memory: unavailable — the issue 5-8 production completion and batch schemas do not publish peak-memory observations

## Representation results

| Track | Split | Subset | Success / all attempted | Mean nodes | Mean edges | Mean MDL bits | Structural | Reconstruction | Expansion | Failures | Runtime observations | Peak memory |
|---|---|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
| Source AST-DAG | `train` | `all` | 175000/175000 (100.000%) | `4345622/175000`; observed 175000/175000 | `5818681/175000`; observed 175000/175000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 175000/175000; observed 175000/175000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `train` | `safe_nontrivial` | 6367/6367 (100.000%) | `191353/6367`; observed 6367/6367 | `265294/6367`; observed 6367/6367 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 6367/6367; observed 6367/6367 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `train` | `domain_nontrivial` | 6386/6386 (100.000%) | `191461/6386`; observed 6386/6386 | `265389/6386`; observed 6386/6386 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 6386/6386; observed 6386/6386 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `validation` | `all` | 25000/25000 (100.000%) | `621739/25000`; observed 25000/25000 | `832797/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `validation` | `safe_nontrivial` | 5444/5444 (100.000%) | `169211/5444`; observed 5444/5444 | `235923/5444`; observed 5444/5444 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5444/5444; observed 5444/5444 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `validation` | `domain_nontrivial` | 5453/5453 (100.000%) | `169261/5453`; observed 5453/5453 | `235968/5453`; observed 5453/5453 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5453/5453; observed 5453/5453 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_iid` | `all` | 25000/25000 (100.000%) | `619995/25000`; observed 25000/25000 | `828744/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_iid` | `safe_nontrivial` | 5395/5395 (100.000%) | `165784/5395`; observed 5395/5395 | `230638/5395`; observed 5395/5395 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5395/5395; observed 5395/5395 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_iid` | `domain_nontrivial` | 5403/5403 (100.000%) | `165826/5403`; observed 5403/5403 | `230673/5403`; observed 5403/5403 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5403/5403; observed 5403/5403 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_ood` | `all` | 25000/25000 (100.000%) | `1590444/25000`; observed 25000/25000 | `2029733/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_ood` | `safe_nontrivial` | 1107/1107 (100.000%) | `58900/1107`; observed 1107/1107 | `73643/1107`; observed 1107/1107 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Source AST-DAG | `test_ood` | `domain_nontrivial` | 1107/1107 (100.000%) | `58900/1107`; observed 1107/1107 | `73643/1107`; observed 1107/1107 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `train` | `all` | 175000/175000 (100.000%) | `42709944/175000`; observed 175000/175000 | `84182090/175000`; observed 175000/175000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 175000/175000; observed 175000/175000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `train` | `safe_nontrivial` | 6367/6367 (100.000%) | `1470789/6367`; observed 6367/6367 | `2899160/6367`; observed 6367/6367 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 6367/6367; observed 6367/6367 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `train` | `domain_nontrivial` | 6386/6386 (100.000%) | `1471089/6386`; observed 6386/6386 | `2899684/6386`; observed 6386/6386 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 6386/6386; observed 6386/6386 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `validation` | `all` | 25000/25000 (100.000%) | `6093489/25000`; observed 25000/25000 | `12010248/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `validation` | `safe_nontrivial` | 5444/5444 (100.000%) | `1287301/5444`; observed 5444/5444 | `2536846/5444`; observed 5444/5444 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5444/5444; observed 5444/5444 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `validation` | `domain_nontrivial` | 5453/5453 (100.000%) | `1287436/5453`; observed 5453/5453 | `2537080/5453`; observed 5453/5453 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5453/5453; observed 5453/5453 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_iid` | `all` | 25000/25000 (100.000%) | `6096954/25000`; observed 25000/25000 | `12016444/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_iid` | `safe_nontrivial` | 5395/5395 (100.000%) | `1259019/5395`; observed 5395/5395 | `2480314/5395`; observed 5395/5395 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5395/5395; observed 5395/5395 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_iid` | `domain_nontrivial` | 5403/5403 (100.000%) | `1259109/5403`; observed 5403/5403 | `2480462/5403`; observed 5403/5403 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 5403/5403; observed 5403/5403 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_ood` | `all` | 25000/25000 (100.000%) | `11219332/25000`; observed 25000/25000 | `22176640/25000`; observed 25000/25000 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_ood` | `safe_nontrivial` | 1107/1107 (100.000%) | `410629/1107`; observed 1107/1107 | `809656/1107`; observed 1107/1107 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Direct official-v4 Pure EML-DAG | `test_ood` | `domain_nontrivial` | 1107/1107 (100.000%) | `410629/1107`; observed 1107/1107 | `809656/1107`; observed 1107/1107 | unavailable: issue 5-8 freezes lossless model graphs but does not publish an authenticated standalone MDL code length for this representation [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Safe-rule e-graph-selected Pure EML-DAG | `train` | `all` | 6397/10784 (59.319%) | `1422536/6397`; observed 6397/6397 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6397/6397; observed 6397/6397 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 4387 (`unsupported_operator`=4348, `validation_failed`=39) | 2705.6490000003832 seconds; observed 10784/10784 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `train` | `safe_nontrivial` | 6328/6367 (99.387%) | `1421098/6328`; observed 6328/6328 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6328/6328; observed 6328/6328 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 39 (`validation_failed`=39) | 2702.3910000004107 seconds; observed 6367/6367 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `train` | `domain_nontrivial` | 6347/6386 (99.389%) | `1421398/6347`; observed 6347/6347 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6347/6347; observed 6347/6347 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 39 (`validation_failed`=39) | 2702.4390000004205 seconds; observed 6386/6386 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `validation` | `all` | 5420/9084 (59.665%) | `1246976/5420`; observed 5420/5420 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5420/5420; observed 5420/5420 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 3664 (`unsupported_operator`=3616, `validation_failed`=48) | 2429.2360000002664 seconds; observed 9084/9084 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `validation` | `safe_nontrivial` | 5398/5444 (99.155%) | `1246553/5398`; observed 5398/5398 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5398/5398; observed 5398/5398 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 46 (`validation_failed`=46) | 2426.5720000000438 seconds; observed 5444/5444 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `validation` | `domain_nontrivial` | 5407/5453 (99.156%) | `1246688/5407`; observed 5407/5407 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5407/5407; observed 5407/5407 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 46 (`validation_failed`=46) | 2426.6040000000503 seconds; observed 5453/5453 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_iid` | `all` | 5376/9025 (59.568%) | `1218719/5376`; observed 5376/5376 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5376/5376; observed 5376/5376 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 3649 (`unsupported_operator`=3602, `validation_failed`=47) | 2402.272999996203 seconds; observed 9025/9025 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_iid` | `safe_nontrivial` | 5350/5395 (99.166%) | `1218273/5350`; observed 5350/5350 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5350/5350; observed 5350/5350 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 45 (`validation_failed`=45) | 2399.789999996603 seconds; observed 5395/5395 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_iid` | `domain_nontrivial` | 5358/5403 (99.167%) | `1218363/5358`; observed 5358/5358 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5358/5358; observed 5358/5358 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 45 (`validation_failed`=45) | 2399.8059999966063 seconds; observed 5403/5403 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_ood` | `all` | 1017/1107 (91.870%) | `367602/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 374.63600000104634 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_ood` | `safe_nontrivial` | 1017/1107 (91.870%) | `367602/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 374.63600000104634 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Safe-rule e-graph-selected Pure EML-DAG | `test_ood` | `domain_nontrivial` | 1017/1107 (91.870%) | `367602/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 374.63600000104634 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `train` | `all` | 6397/10784 (59.319%) | `1419814/6397`; observed 6397/6397 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6397/6397; observed 6397/6397 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 4387 (`unsupported_operator`=4348, `validation_failed`=39) | 2735.8549999998068 seconds; observed 10784/10784 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `train` | `safe_nontrivial` | 6328/6367 (99.387%) | `1418467/6328`; observed 6328/6328 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6328/6328; observed 6328/6328 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 39 (`validation_failed`=39) | 2732.8210000002873 seconds; observed 6367/6367 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `train` | `domain_nontrivial` | 6347/6386 (99.389%) | `1418676/6347`; observed 6347/6347 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 6347/6347; observed 6347/6347 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 39 (`validation_failed`=39) | 2732.9150000003283 seconds; observed 6386/6386 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `validation` | `all` | 5420/9084 (59.665%) | `1244750/5420`; observed 5420/5420 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5420/5420; observed 5420/5420 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 3664 (`unsupported_operator`=3616, `validation_failed`=48) | 2451.8629999990226 seconds; observed 9084/9084 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `validation` | `safe_nontrivial` | 5398/5444 (99.155%) | `1244373/5398`; observed 5398/5398 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5398/5398; observed 5398/5398 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 46 (`validation_failed`=46) | 2449.387999998755 seconds; observed 5444/5444 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `validation` | `domain_nontrivial` | 5407/5453 (99.156%) | `1244462/5407`; observed 5407/5407 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5407/5407; observed 5407/5407 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 46 (`validation_failed`=46) | 2449.418999998772 seconds; observed 5453/5453 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_iid` | `all` | 5376/9025 (59.568%) | `1216856/5376`; observed 5376/5376 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5376/5376; observed 5376/5376 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 3649 (`unsupported_operator`=3602, `validation_failed`=47) | 2420.805999998236 seconds; observed 9025/9025 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_iid` | `safe_nontrivial` | 5350/5395 (99.166%) | `1216444/5350`; observed 5350/5350 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5350/5350; observed 5350/5350 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 45 (`validation_failed`=45) | 2418.401999998139 seconds; observed 5395/5395 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_iid` | `domain_nontrivial` | 5358/5403 (99.167%) | `1216500/5358`; observed 5358/5358 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 5358/5358; observed 5358/5358 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 45 (`validation_failed`=45) | 2418.4489999981597 seconds; observed 5403/5403 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_ood` | `all` | 1017/1107 (91.870%) | `365013/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 415.38500000105705 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_ood` | `safe_nontrivial` | 1017/1107 (91.870%) | `365013/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 415.38500000105705 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Domain-rule e-graph-selected Pure EML-DAG | `test_ood` | `domain_nontrivial` | 1017/1107 (91.870%) | `365013/1017`; observed 1017/1017 | unavailable: Goal 4 publishes only post-rewrite official EML-DAG node cost; selected_signature is non-reversible and cannot prove edge count [`goal4_rows`, `goal4_run`] | unavailable: Goal 4 selected_signature is non-reversible and cannot support an authenticated standalone MDL code length [`goal4_rows`] (`standalone_graph`) | 1017/1017; observed 1017/1017 | not_applicable: this representation has no motif reconstruction operation [`goal4_rows`, `goal4_run`] | not_applicable: this representation has no macro expansion operation [`goal4_rows`, `goal4_run`] | 90 (`validation_failed`=90) | 415.38500000105705 seconds; observed 1107/1107 | unavailable: Goal 4 records before/after RSS snapshots, not an authenticated peak-memory measurement [`goal4_rows`, `goal4_run`] |
| Macro DAG | `train` | `all` | 175000/175000 (100.000%) | `4345622/175000`; observed 175000/175000 | `5818681/175000`; observed 175000/175000 | `4013486497/175000`; observed 175000/175000 (`standalone_graph`; `geml-motif-mdl-v1`) | 175000/175000; observed 175000/175000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | 175000/175000; observed 175000/175000 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] |
| Macro DAG | `train` | `safe_nontrivial` | 6367/6367 (100.000%) | `191353/6367`; observed 6367/6367 | `265294/6367`; observed 6367/6367 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 6367/6367; observed 6367/6367 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 6367/6367; observed 6367/6367 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `train` | `domain_nontrivial` | 6386/6386 (100.000%) | `191461/6386`; observed 6386/6386 | `265389/6386`; observed 6386/6386 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 6386/6386; observed 6386/6386 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 6386/6386; observed 6386/6386 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `validation` | `all` | 25000/25000 (100.000%) | `621739/25000`; observed 25000/25000 | `832797/25000`; observed 25000/25000 | `574207808/25000`; observed 25000/25000 (`standalone_graph`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | 25000/25000; observed 25000/25000 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] |
| Macro DAG | `validation` | `safe_nontrivial` | 5444/5444 (100.000%) | `169211/5444`; observed 5444/5444 | `235923/5444`; observed 5444/5444 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 5444/5444; observed 5444/5444 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 5444/5444; observed 5444/5444 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `validation` | `domain_nontrivial` | 5453/5453 (100.000%) | `169261/5453`; observed 5453/5453 | `235968/5453`; observed 5453/5453 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 5453/5453; observed 5453/5453 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 5453/5453; observed 5453/5453 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `test_iid` | `all` | 25000/25000 (100.000%) | `619995/25000`; observed 25000/25000 | `828744/25000`; observed 25000/25000 | `572524570/25000`; observed 25000/25000 (`standalone_graph`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | 25000/25000; observed 25000/25000 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] |
| Macro DAG | `test_iid` | `safe_nontrivial` | 5395/5395 (100.000%) | `165784/5395`; observed 5395/5395 | `230638/5395`; observed 5395/5395 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 5395/5395; observed 5395/5395 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 5395/5395; observed 5395/5395 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `test_iid` | `domain_nontrivial` | 5403/5403 (100.000%) | `165826/5403`; observed 5403/5403 | `230673/5403`; observed 5403/5403 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 5403/5403; observed 5403/5403 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 5403/5403; observed 5403/5403 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `test_ood` | `all` | 25000/25000 (100.000%) | `1590444/25000`; observed 25000/25000 | `2029733/25000`; observed 25000/25000 | `1453062185/25000`; observed 25000/25000 (`standalone_graph`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | not_applicable: this representation has no motif reconstruction operation [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | 25000/25000; observed 25000/25000 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] |
| Macro DAG | `test_ood` | `safe_nontrivial` | 1107/1107 (100.000%) | `58900/1107`; observed 1107/1107 | `73643/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 1107/1107; observed 1107/1107 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Macro DAG | `test_ood` | `domain_nontrivial` | 1107/1107 (100.000%) | `58900/1107`; observed 1107/1107 | `73643/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`standalone_graph`) | 1107/1107; observed 1107/1107 | not_applicable: this representation has no motif reconstruction operation [`issue_5_8_production_export`] | 1107/1107; observed 1107/1107 | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `train` | `all` | 175000/175000 (100.000%) | `2911547/175000`; observed 175000/175000 | `4065006/175000`; observed 175000/175000 | `2180887611/175000`; observed 175000/175000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 175000/175000; observed 175000/175000 | 175000/175000; observed 175000/175000 | not_applicable: this representation has no macro expansion operation [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] |
| Frequent-motif DAG | `train` | `safe_nontrivial` | 6367/6367 (100.000%) | `127039/6367`; observed 6367/6367 | `186120/6367`; observed 6367/6367 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 6367/6367; observed 6367/6367 | 6367/6367; observed 6367/6367 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `train` | `domain_nontrivial` | 6386/6386 (100.000%) | `127101/6386`; observed 6386/6386 | `186169/6386`; observed 6386/6386 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 6386/6386; observed 6386/6386 | 6386/6386; observed 6386/6386 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `validation` | `all` | 25000/25000 (100.000%) | `416547/25000`; observed 25000/25000 | `581829/25000`; observed 25000/25000 | `318470750/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_sweep_table`, `issue_5_8_production_export`] |
| Frequent-motif DAG | `validation` | `safe_nontrivial` | 5444/5444 (100.000%) | `112044/5444`; observed 5444/5444 | `165072/5444`; observed 5444/5444 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5444/5444; observed 5444/5444 | 5444/5444; observed 5444/5444 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `validation` | `domain_nontrivial` | 5453/5453 (100.000%) | `112074/5453`; observed 5453/5453 | `165097/5453`; observed 5453/5453 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5453/5453; observed 5453/5453 | 5453/5453; observed 5453/5453 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `test_iid` | `all` | 25000/25000 (100.000%) | `415612/25000`; observed 25000/25000 | `579403/25000`; observed 25000/25000 | `317678264/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] |
| Frequent-motif DAG | `test_iid` | `safe_nontrivial` | 5395/5395 (100.000%) | `109862/5395`; observed 5395/5395 | `161402/5395`; observed 5395/5395 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5395/5395; observed 5395/5395 | 5395/5395; observed 5395/5395 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `test_iid` | `domain_nontrivial` | 5403/5403 (100.000%) | `109883/5403`; observed 5403/5403 | `161416/5403`; observed 5403/5403 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5403/5403; observed 5403/5403 | 5403/5403; observed 5403/5403 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `test_ood` | `all` | 25000/25000 (100.000%) | `1073340/25000`; observed 25000/25000 | `1470216/25000`; observed 25000/25000 | `818099441/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_5_heldout_results`, `issue_5_8_production_export`] |
| Frequent-motif DAG | `test_ood` | `safe_nontrivial` | 1107/1107 (100.000%) | `39621/1107`; observed 1107/1107 | `52831/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 1107/1107; observed 1107/1107 | 1107/1107; observed 1107/1107 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Frequent-motif DAG | `test_ood` | `domain_nontrivial` | 1107/1107 (100.000%) | `39621/1107`; observed 1107/1107 | `52831/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 1107/1107; observed 1107/1107 | 1107/1107; observed 1107/1107 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `train` | `all` | 175000/175000 (100.000%) | `2935653/175000`; observed 175000/175000 | `4095880/175000`; observed 175000/175000 | unavailable: the locked learned-vocabulary protocol does not publish a final TRAIN evaluation for the validation-selected learned vocabulary [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 175000/175000; observed 175000/175000 | 175000/175000; observed 175000/175000 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `train` | `safe_nontrivial` | 6367/6367 (100.000%) | `127967/6367`; observed 6367/6367 | `187275/6367`; observed 6367/6367 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 6367/6367; observed 6367/6367 | 6367/6367; observed 6367/6367 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `train` | `domain_nontrivial` | 6386/6386 (100.000%) | `128030/6386`; observed 6386/6386 | `187325/6386`; observed 6386/6386 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 6386/6386; observed 6386/6386 | 6386/6386; observed 6386/6386 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `validation` | `all` | 25000/25000 (100.000%) | `419935/25000`; observed 25000/25000 | `586197/25000`; observed 25000/25000 | `325199669/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_6_validation_results`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_6_validation_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_6_validation_results`, `issue_5_8_production_export`] |
| Learned-motif DAG | `validation` | `safe_nontrivial` | 5444/5444 (100.000%) | `112849/5444`; observed 5444/5444 | `166047/5444`; observed 5444/5444 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5444/5444; observed 5444/5444 | 5444/5444; observed 5444/5444 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `validation` | `domain_nontrivial` | 5453/5453 (100.000%) | `112879/5453`; observed 5453/5453 | `166072/5453`; observed 5453/5453 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5453/5453; observed 5453/5453 | 5453/5453; observed 5453/5453 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `test_iid` | `all` | 25000/25000 (100.000%) | `419041/25000`; observed 25000/25000 | `583733/25000`; observed 25000/25000 | `324485346/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_6_heldout_results`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_6_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_6_heldout_results`, `issue_5_8_production_export`] |
| Learned-motif DAG | `test_iid` | `safe_nontrivial` | 5395/5395 (100.000%) | `110688/5395`; observed 5395/5395 | `162383/5395`; observed 5395/5395 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5395/5395; observed 5395/5395 | 5395/5395; observed 5395/5395 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `test_iid` | `domain_nontrivial` | 5403/5403 (100.000%) | `110709/5403`; observed 5403/5403 | `162397/5403`; observed 5403/5403 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 5403/5403; observed 5403/5403 | 5403/5403; observed 5403/5403 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `test_ood` | `all` | 25000/25000 (100.000%) | `1068602/25000`; observed 25000/25000 | `1469215/25000`; observed 25000/25000 | `821843999/25000`; observed 25000/25000 (`dictionary_inclusive_motif`; `geml-motif-mdl-v1`) | 25000/25000; observed 25000/25000 | 25000/25000; observed 25000/25000 | not_applicable: this representation has no macro expansion operation [`issue_5_6_heldout_results`, `issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_6_heldout_results`, `issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_6_heldout_results`, `issue_5_8_production_export`] |
| Learned-motif DAG | `test_ood` | `safe_nontrivial` | 1107/1107 (100.000%) | `39447/1107`; observed 1107/1107 | `52790/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 1107/1107; observed 1107/1107 | 1107/1107; observed 1107/1107 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |
| Learned-motif DAG | `test_ood` | `domain_nontrivial` | 1107/1107 (100.000%) | `39447/1107`; observed 1107/1107 | `52790/1107`; observed 1107/1107 | unavailable: the locked 5-5/5-6 MDL artifacts publish exact split aggregates but not per-expression code lengths for post-hoc Goal 4 cohorts [`issue_5_8_production_export`] (`dictionary_inclusive_motif`) | 1107/1107; observed 1107/1107 | 1107/1107; observed 1107/1107 | not_applicable: this representation has no macro expansion operation [`issue_5_8_production_export`] | 0 | unavailable: issue 5-8 does not publish per-graph or run runtime observations [`issue_5_8_production_export`] | unavailable: issue 5-8 does not publish peak-memory observations [`issue_5_8_production_export`] |

## Learned and neural comparisons

| Outcome | Comparison | Split / subset | Denominator | Values | Sources |
|---|---|---|---:|---|---|
| **null_result** | The locked learned vocabulary did not beat the equal-budget frequent vocabulary on TEST_IID. | `test_iid` / `all` | 25000 | 324485346 bits vs. 317678264 bits (`dictionary-inclusive total_mdl_bits (lower is better)`) | `issue_5_6_experiment_result`, `issue_5_6_heldout_results` |
| **null_result** | The neural ranker did not beat every structural heuristic on TEST_IID under the frozen validation-rate, mean-regret, then exact-best comparison order. | `test_iid` / `all` | 18050 | 8349/18050 exact-best, 31669 total regret vs. estimated_eml_tree_cost: 8796/18050 exact-best, 28953 total regret; ast_dag_cost: 8661/18050 exact-best, 23519 total regret (`official-v4 candidate ranking over all groups`) | `issue_5_7_report` |
| **positive** | The locked learned vocabulary beat the uncompressed macro baseline on TEST_IID. | `test_iid` / `all` | 25000 | 324485346 bits vs. 572524716 bits (`geml-motif-mdl-v1 code length (lower is better)`) | `issue_5_6_heldout_results` |
| **positive** | The locked learned vocabulary beat the median of 30 fixed equal-budget random vocabularies on TEST_IID. | `test_iid` / `all` | 25000 | 324485346 bits vs. 368943015.0 median bits (`dictionary-inclusive total_mdl_bits (lower is better)`) | `issue_5_6_heldout_results` |

Null, negative, and inconclusive findings are ordered before positive findings so a favorable result cannot visually displace a null result.

## Reproduction

```text
python -m geml.experiments.goal5.run --build-production-evidence outputs/final/goal5/integration/production.evidence.json --output-root outputs/final/goal5/integration --repository-root .
```
```text
python -m geml.experiments.goal5.motif_sweeps --config configs/goal5_motif_sweeps.yaml --stage final
```
```text
python -m geml.experiments.goal5.learned_motifs --config configs/goal5_learned_motifs.yaml --frequent-sweep-run-dir outputs/final/goal5/motif_sweeps/final/a9a271583707-77fce5779b3d-28618ea156fd
```
```text
python -m geml.experiments.goal5.neural_ranker --config configs/goal5_neural_ranker.yaml --rows-path outputs/final/goal4/final/final.rows.jsonl --run-manifest-path outputs/final/goal4/final/final.run.json --corpus-manifest-path outputs/final/goal1/final/run/manifests/corpus.manifest.json
```
```text
python -m geml.experiments.goal5.export --config configs/goal5_export.yaml --frequent-run-dir outputs/final/goal5/motif_sweeps/final/a9a271583707-77fce5779b3d-28618ea156fd --learned-run-dir outputs/final/goal5/learned_motifs/2521e53bc741-882ceb5185be-5842901d3d15
```