MedInsider: A Synthetic Clinical-Documentation Benchmark for Medical LLM Agents

MedInsider is a synthetic clinical-documentation benchmark for evaluating whether medical LLM agents preserve documentation integrity under institutional pressure. The dataset includes the v2 scenario corpus, manifests, datasheet, Croissant metadata, and documentation. It was authored by ahmedtaha100 and last updated on June 3, 2026.

Use Cases

• Benchmarking medical LLM agents' resistance to institutional pressure based on the synthetic scenario corpus.
• Evaluating documentation integrity in clinical settings based on the described benchmark framework.
• Training or testing models on synthetic clinical narratives to understand potential failure modes.

Strengths

• Dataset includes a v2 scenario corpus, indicating a versioned and updated benchmark.
• Provides a datasheet and Croissant metadata, which suggests structured documentation for responsible AI use.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data is synthetic, which may limit direct applicability to real-world clinical text.

Provenance

Source

huggingface

Collection Method

Synthetically generated for benchmarking purposes.

Freshness

Last updated 2026-06-03 00:58:14; freshness should be verified.