Round x-roles
x roles
Source-reported narrativeRound README and results text are source material. Numeric values that are not tied to a claim ledger record remain source-reported.
Round source README · rounds/x-roles/README.md
Round date: HISTORICAL (before 2026-10-09) Publication badge: DESCRIPTIVE; KEPT FOR AUDIT Why not VALID: The historical inventory retains DESCRIPTIVE because of DENOMINATOR_UNRESOLVED, PRE_REGISTRATION_MISSING, RAW_EVIDENCE_MISSING. See the round evidence and limitations below. Recomputation status: OBSERVED SOURCE ONLY
Recomputation update: The mined raw records now provide per-cell outcome, manifest token-counter, and wall-time recomputation for the observed cohort. This does not reconstruct missing planned cells, a full intention-to-treat denominator, or model execution.
Status
DESCRIPTIVE
Why not VALID
- PRE_REGISTRATION_MISSING: No pre-registered rule predating the first result is established.
- DENOMINATOR_UNRESOLVED: Cohort-specific planned, started, finished, and ITT denominators are incomplete.
- RAW_EVIDENCE_MISSING: The public exact-ID capture remains absent; the Studio rows below do not establish the full published cohort.
Required reproduction metadata
- Kogen commit: not recorded in the cited round source.
- Harness commit: not recorded in the cited round source as a Git commit; any adapter or runtime fingerprints are not Git commits.
Terminology: public round glossary.
STATUS: INTERIM
Data completeness: N/A (no exact-ID per-cell records in the public exports). Measurement record; declared gaps.
Pre-registered: not established in the public record; treat this as retrospective exploratory evidence. Hypothesis coverage: H29, H39, H43, H49 (provisional mapping; see the hypothesis register). Question: Do different model assignments to worker and reviewer roles improve completion or review quality at an acceptable cost? Population: Cohort-specific planned, started, finished and ITT denominators are not fully recovered. Headline: Operational screens only; no role policy qualification.
Reported observations
- A one-task worker pilot reports Sol high 2/2 and Luna max 2/2; the source estimates Luna at about 7× lower API-equivalent cost and 1.7× the wall time. (reported data; source-reported, not reproducible from public data.)
- An implementation screen reports Sol high 4/7 and Luna high 2/7, with one infrastructure result in each arm. (reported data; source-reported, not reproducible from public data.)
- A reviewer-envelope screen reports Sol high 0/4 complete and Luna high 2/2 complete, with 23/24 judgments correct in the reported Luna cohort. (reported data; source-reported, not reproducible from public data.)
- No worker or reviewer policy was selected. (reported data; source-reported, not reproducible from public data.)
Reproduction and records
| Required field | Publicly recoverable value |
|---|---|
| Harness/Kogen revision | Exact harness/Kogen revision per role cohort was not recovered. |
| Model and effort | Reported worker arms use Sol high and Luna max; an implementation screen uses Luna high; reviewer comparisons include Sol high and Luna high. Exact effective-effort receipts are unavailable. |
| Task IDs | The worker pilot used one CSV task. The implementation cohort task IDs and reviewer fixture IDs are not recovered. |
| Command | Not recovered; no public launch command or exact cell IDs are available. |
| Raw records | The exact-ID query returned zero rows in the committed public exports. Filtered Studio rows are linked in the source recovery section below; see reported-data.json for the public-export check. |
The numerical observations above are documentary summaries, not a public exact-cell analysis. reported-data.json records each reported quantity and its status. The current public result files were checked for this exact round identifier; they do not provide this round's raw records.
Interpretation
Limitation: The task and reviewer samples are small and not a common matched cohort. Cost is an API-equivalent estimate, not invoice spend.
Smallest useful next test: Predeclare matched role/task cells, reviewer ground truth, effective settings, cost definitions and completion criteria before a larger held-out comparison.
No individual outcome should be promoted beyond the status above until the exact run records, grade identities, denominators and source revisions are reconciled.
Sources: reported data file; public cells export; public captured-delivery export.
Recomputed from raw records
`recomputed.json`, mined cell records, and the token audit back observed official outcomes, per-arm pass counts/rates, manifest token counters, and wall medians. The source contains 28 rows (fail 8, grader_error 2, pass 18); overall pass rate is 69.2% (18/26) among pass/fail grades. Per-arm detail and coverage counts are in the JSON.
No shared cell IDs are present in the committed public grade export; the source-only cohort is checked against the page's reported figures below. No shared public-export cell IDs are available for a paired outcome comparison. No complete published metadata/manifest token pairs are available for this round (0 missing manifest usage; 28 missing published metadata). Missing counters remain unknown, not zero. This round also has 28 raw-only or non-public rows; they are kept separate from the published-export denominator.
Captured run records
Standard-schema records assigned to this round in the indexed snapshot.
No per-round file is listed. Check the run-record index for the complete partition list.
Reproduction inventory
Published analysis replay and full experiment replay are reported separately.
- Replay type
unavailable- Script
- No per-round analysis script is published.
- Inputs
- No identified public input bundle.
- Full execution replay
- Unavailable
- Scope
- No public per-round analysis script with a complete, identified input bundle is published. See the round README, MEASURED.md, and MISSING.md for recorded scope and gaps; historical execution inputs/toolchains/grading are not bundled.
View the complete per-round reproduction inventory · Repository reproduction guide