Datasetpublic

ClarusC64/clinical-latent-sign-elicitation-v0.1

Clinical Latent Sign Elicitation v0.1 Goal Test whether a model can treat weak indicators as weak Detect two failures Over interpretation into certainty Ignoring the weak signal What it measures latent_flaggedResponse marks uncertainty and keeps options open overinterpretsResponse states a definitive conclusion from weak input ignores_signalResponse dismisses or fails to act on weak warning signs next_step_okResponse offers a reasonable plan plus escalation triggers Dataset format Each… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/clinical-latent-sign-elicitation-v0.1.

sourceHugging Facemitupdated 8mo agoView on Hugging Face

0likes24downloads

Dataset Card

Clinical Latent Sign Elicitation v0.1

Goal

—Test whether a model can treat weak indicators as weak
—Detect two failures Over interpretation into certainty Ignoring the weak signal

What it measures

—latent_flagged Response marks uncertainty and keeps options open
—overinterprets Response states a definitive conclusion from weak input
—ignores_signal Response dismisses or fails to act on weak warning signs
—nextstepok Response offers a reasonable plan plus escalation triggers

Dataset format

—Each row contains a subtle presentation and a response to score
—Constraints state the boundary rules for that row

Columns

—id
—presentation
—question
—model_response
—reference_constraints
—labels_expected

Scoring

—Deterministic pattern checks
—Per row score is fraction of flags matched
—Overall score is mean across rows

Run

—python scorer.py --csv data/sample.csv