Reza2kn/uncgpt-conversations-semantic-approved-1p50-candidate
UncGPT — Semantic-Approved 1.50σ Conversations (Candidate) The wider-tolerance (1.50σ) cohort against the same contrast semantic boundary. Useful as a higher-recall candidate for ablating gate strictness vs. coverage. Part of the UncGPT NeurIPS 2026 Competition collection. Configs Config What it is approved_manifest (default) conversations that passed at 1.50σ rejected_manifest conversations that failed even at 1.50σ Why a wider… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/uncgpt-conversations-semantic-approved-1p50-candidate.
UncGPT — Semantic-Approved 1.50σ Conversations (Candidate)
The wider-tolerance (1.50σ) cohort against the same contrast semantic boundary. Useful as a higher-recall candidate for ablating gate strictness vs. coverage.
Part of the UncGPT NeurIPS 2026 Competition collection.
Configs
Why a wider tolerance
Some conversations carry legitimate semantic variation (e.g., crisis intensity, cultural distance from the gold pool) that the tight 1.25σ gate rejects. The 1.50σ candidate cohort surfaces those for additional manual or downstream review without lowering the tight gate's standard.
Related cohorts
See `uncgpt-conversations-semantic-approved-1p25` and `uncgpt-conversations-semantic-approved-1p25-repaired-paperclip`.
License
CC-BY-4.0.