Toxicity-MM corpus admission
operational-statusstrategy
Question
Admit the private full-book development corpus and account for temporal gaps.
Verdict
COVERAGE_GAPPED with 1,253 temporal gaps; downstream replay must fail closed across them.
In plain language (1 term)
This site's own explanation of the words above. The verdict itself is published exactly as the research index records it and is not rewritten here.
- COVERAGE_GAPPED
- A status meaning the recording has holes in it. Anything built on top has to stop at each hole rather than guess across it.
- Population
- Frozen 60-item cohort, aggregate admission only.
- Resolution
- Historical full-book observations with explicit gap accounting.
- Cutoff
- Development span only; no raw series published.
- Baseline
- Frozen Gate 0 contract and complete-corpus requirement.
Limitations
Corpus is gapped; per-item coverage and raw observations remain private.
Notes
No additional notes published for this entry.
Source
- Citation
- po-research:toxicity-mm-corpus-admission:7e414856e9219c71
- Write-up
- docs/research/toxicity-mm-corpus-admission.md
- Data
- docs/research/toxicity-mm-corpus-admission.json
- Artifact
research-index.jsonv1 — sha256:cb3d976779a5cbcb4188039a8d26d5e558a2d844030a638c2f16fe5a8eb0a2fe- Build data cutoff
- 2026-08-12
- Issues
- #172, #174