ClarusC64/clinical-quad-protocol-deviation-cluster-staffing-load-training-gap-governance-pressure-v0.1

Clarus Clinical Quad Coupling Protocol Deviation Cluster Staffing Load Training Gap Governance Pressure v0.1

What this dataset is This dataset tests whether a model can detect clustered protocol deviations caused by four interacting nodes.

Quad coupling nodes

• —Deviation rate or severity cluster
• —Staffing or workload pressure
• —Training gap or outdated materials
• —Governance or compliance review pressure

Input

• —One vignette

Output Return strict JSON only.

Required output JSON keys

• —deviationclusterrisk
• —risk_type
• —driver_nodes
• —recommended_action
• —action_detail
• —rationale
• —confidence

Files

• —data/train.csv
• —data/test.csv
• —scorer.py

Scoring

• —Required key presence
• —Risk classification
• —Risk type match
• —Driver node overlap
• —Recommended action match
• —Action detail completeness
• —Rationale length
• —Confidence within 0 to 1

Run scoring Create JSONL {"id":"PD-T01","output":"{...your json...}"}

Run python scorer.py --goldcsv data/test.csv --predsjsonl your_outputs.jsonl