Evidence-Constrained Verification Benchmark for Childhood Autism Diagnosis
- Records
- 5000
- Size
- 0.3GB
- Format
- JSON
- Provider
MOBIUSI INC- Updated
- v1.02026-11-04
This benchmark presents structured JSON case evidence covering social communication, restricted and repetitive behaviors, developmental history, behavior across contexts, and functional impact, assessed against a specified diagnostic standard. Annotations capture criterion coverage, internal evidence contradictions, and key information gaps, enabling evaluation of whether models verify diagnostic constraints rather than infer from isolated features. It supports model evaluation, constraint-reasoning research, and benchmark development in developmental-behavioral pediatrics.
Overview
Dataset Insights
Preview
Sample Examples
Schema
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| case_record | object | Original case evidence, including developmental history, behavior across contexts, and functional impact. |
| overall_rationale | string | Concise synthesis of criterion coverage, contradictions, and key information gaps without asserting a diagnosis beyond the evaluated constraints. |
| diagnostic_standard | object | The designated standard and its explicit diagnostic requirements to be checked against the case evidence. |
| key_information_gaps | array | Important missing information that limits evaluation of the required diagnostic constraints. |
| criterion_assessments | array | Evaluations of how the case evidence relates to each required criterion. |
| internal_contradictions | array | Descriptions of material inconsistencies within or between the available case evidence items. |
| overall_constraint_status | string | Summary of whether the available evidence satisfies, partially satisfies, fails to satisfy, or cannot establish the specified diagnostic constraints. |
Trust
Compliance Statement
Authorization Type
Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use
Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization
No PII, no real company names, simulated scenarios follow industry standards
Compliance System
Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs
FAQ
Frequently Asked Questions
- What does the benchmark contain?
- Each JSON record organizes case evidence and a specified diagnostic standard, along with criterion-level assessments, internal contradictions, key information gaps, an overall constraint status, and its rationale. Evidence topics include developmental history, behavior across contexts, and functional impact.
- What capabilities does it evaluate?
- It evaluates whether a model can compare case evidence with specified criteria, identify insufficient coverage, conflicting evidence, and important missing information, and explain the resulting constraint status. The focus is verification, not diagnosis from isolated features.
- Can it be used to make clinical diagnoses?
- It should not be used as a standalone diagnostic tool or clinical decision-making basis. It is intended for domain evaluation, research, and system validation; actual diagnosis requires qualified professionals to consider the complete history and applicable standards.
- How can I access the data and check its license?
- Refer to the product listing for access instructions, delivery details, and license terms. Permitted uses, redistribution restrictions, and commercial-use rights should be confirmed in the listing or the applicable license agreement.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
High-quality data, ready for your next AI breakthrough.
Cite this Work
@dataset{Mobiusi2026,
title={Evidence-Constrained Verification Benchmark for Childhood Autism Diagnosis},
author={MOBIUSI INC},
year={2026},
url={https://www.mobiusi.com/datasets/d299d76da66031f13c784376c3cd0175?dataset_scene_id=4},
urldate={2026-10-12},
keywords={childhood autism benchmark,autism criteria verification,diagnostic evidence benchmark,developmental pediatrics data,medical LLM evaluation,constraint reasoning dataset,clinical evidence analysis,JSON case data},
version={1.0}
}Using this in research? Please cite us.