Clinical Differential Narrative Coherence Scoring V0.1

Clinical Differential Narrative Coherence Scoring V0.1 is a dataset for evaluating AI models on scoring diagnostic explanations. The dataset, authored by ClarusC64, was last updated on February 2, 2026. It tests whether a model can assess candidate diagnoses based on their explanatory coherence across multiple evidence streams.

Use Cases

• Benchmarking AI model performance on clinical diagnostic reasoning based on the described coherence criteria.
• Training models to generate or score explanatory narratives that link findings into a single mechanism.
• Evaluating how AI models handle contradictory evidence in medical cases without resorting to patchwork explanations.

Strengths

• Dataset has a clearly defined purpose for testing model performance on explanatory coherence.
• The description specifies required outputs: diagnosis_id, coherence_score_0_100, and unexplained_findings.
• Coherence is defined across five evidence streams: imaging, labs, histology, exposure, and course.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Sample data is unavailable, preventing preview of data format and content.

Provenance

Source

huggingface

Collection Method

Method of data gathering is not described.

Time Range

Temporal coverage is not specified.

Freshness

Last updated 2026-02-02 12:04:20; freshness should be verified.

Geography

Spatial coverage is not specified.