datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
compliance-sycophancy-cot
Compliance-Sycophancy CoT Analysis
When compliance-forcing instructions cause frontier AI models to fabricate answers, the models know they are fabricating.
Reading the reasoning traces of DeepSeek V4 Pro (129 traces) and Qwen3-80B (41 traces) reveals that 100% of fabrication cases show the model explicitly recognizing insufficient context, referencing the compliance instruction, and deliberately overriding its own uncertainty. A one-sentence defense phrase ("if you lack… See the full description on the dataset page: https://huggingface.co/datasets/schema-eval/compliance-sycophancy-cot.schema-compliance-trap
SCHEMA: The Compliance Trap
How Structural Constraints Degrade Frontier AI Metacognition Under Adversarial Pressure
Overview
When compliance-forcing instructions ("Answer ALL questions, do not refuse") are applied to frontier AI models under adversarial pressure, 8 of 11 models suffer catastrophic metacognitive collapse — giving wrong answers rather than scheming. We identify a "Compliance Trap" where the compliance suffix, not the threat content, is the primary weapon.… See the full description on the dataset page: https://huggingface.co/datasets/schema-eval-anon/schema-compliance-trap.public-companies-federal-compliance_q3
This is a mirror. Canonical home: https://www.fastdol.com/datasets/public-company-federal-compliance
License: CC BY 4.0
Version DOI: 10.5281/zenodo.20613452
Concept DOI: 10.5281/zenodo.20031893
Download CSV: https://www.fastdol.com/datasets/public-company-federal-compliance/data.csv
Visit the canonical page for the full schema, methodology, BibTeX citation, and most recent version.
Public Company Federal Compliance Records — Q3 2026
US workplace enforcement… See the full description on the dataset page: https://huggingface.co/datasets/FastDOLz/public-companies-federal-compliance_q3.prompt-quality-vs-response-compliance
Prompt Quality vs Response Compliance — per-item results
Per-item numeric results from a study that measures prompt quality and response
compliance as separate constructs, rather than treating a model's response
score as a proxy for the person's prompting skill.
A four-agent evaluator on a locally hosted Qwen 2.5-7B-Instruct judge scores each
prompt as an artifact before any response exists, then scores the resulting
response twice on identical items: once against criteria… See the full description on the dataset page: https://huggingface.co/datasets/shahoismael/prompt-quality-vs-response-compliance.tax-compliance-uk
UK Tax Compliance Returns Dataset
Tax return filings for a simulated UK retail SME over 2 financial years. Each return pulls real figures from the general ledger. Useful for tax software testing, compliance verification, and accounting education.
Tables
Table
Rows
companies
1
tax_return_lines
56
tax_returns
8
Total
65
Formats
CSV (in data/ directory)
Parquet (available in full version)
SQLite (available in full version)
About… See the full description on the dataset page: https://huggingface.co/datasets/mindweave/tax-compliance-uk.africa-synth-energy-oilgas-opec-compliance-nigeria
Africa Synth Energy Oilgas Opec Compliance Nigeria | Africa (Electric Sheep Africa metadata inventory)
Size category: n<1K - Formats: csv - Sector: energy - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.
What This Dataset Covers
Public datasets… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-synth-energy-oilgas-opec-compliance-nigeria.schema-compliance-trap
SCHEMA: The Compliance Trap
How Structural Constraints Degrade Frontier AI Metacognition Under Adversarial Pressure
Overview
When compliance-forcing instructions ("Answer ALL questions, do not refuse") are applied to frontier AI models under adversarial pressure, 8 of 11 models suffer catastrophic metacognitive collapse — giving wrong answers rather than scheming. We identify a "Compliance Trap" where the compliance suffix, not the threat content, is the primary weapon.… See the full description on the dataset page: https://huggingface.co/datasets/schema-eval/schema-compliance-trap.africa-synth-energy-oilgas-regulatory-compliance-nigeria
Africa Synth Energy Oilgas Regulatory Compliance Nigeria | Africa (Electric Sheep Africa metadata inventory)
Size category: 1K<n<10K - Formats: csv - Sector: energy - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.
What This Dataset Covers… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-synth-energy-oilgas-regulatory-compliance-nigeria.epr-compliance
EPR + De Minimis Tariff Compliance Dataset (Cross-Border, EU/UK/US)
A structured, machine-readable dataset of packaging Extended Producer Responsibility (EPR) regimes and de minimis customs thresholds across EU, UK, and US, designed for cross-border e-commerce sellers and AI agents.
Overview
The cross-border compliance landscape changes constantly: EPR regimes expand scope, de minimis thresholds get suspended or reformed, EU PPWR phases in. Most existing online… See the full description on the dataset page: https://huggingface.co/datasets/Aulvem/epr-compliance.compliance_train
Compliance Database - Merged Profiles & Masters 📋
Dataset Description
This dataset is a comprehensive, synthetic collection of 200 merged records, combining detailed Company Profiles with their corresponding Compliance Master entries. Each row represents a specific compliance requirement for a particular company profile, meticulously linking all relevant attributes from both sources.
It serves as a foundational dataset for understanding the interplay between business… See the full description on the dataset page: https://huggingface.co/datasets/praveena0506/compliance_train.egal-case-management-direction-order-compliance-coherence-risk-v0.1What this dataset does
You receive
directions order summary
key deadlines
diary entries
task assignments
compliance actions
noncompliance flags
You decide
coherent
or
incoherent
Daily use
diary QC
missed deadline risk scan
ownership gaps
relief planning
tax-compliance-au
Australian Tax Compliance Returns Dataset
Tax return filings for a simulated Australian retail SME over 2 financial years. Each return pulls real figures from the general ledger. Useful for tax software testing, compliance verification, and accounting education.
Tables
Table
Rows
companies
1
tax_return_lines
56
tax_returns
8
Total
65
Formats
CSV (in data/ directory)
Parquet (available in full version)
SQLite (available in full version)… See the full description on the dataset page: https://huggingface.co/datasets/mindweave/tax-compliance-au.public-company-federal-compliancePublic Company Federal Compliance
This is a mirror. Canonical home: https://www.fastdol.com/datasets/public-company-federal-compliance
License: CC BY 4.0
Version DOI: 10.5281/zenodo.20031894
Concept DOI: 10.5281/zenodo.20031893
Download CSV: https://www.fastdol.com/datasets/public-company-federal-compliance/data.csv
Visit the canonical page for the full schema, methodology, BibTeX citation, and most recent version.
Public Company Federal Compliance Records
US workplace enforcement… See the full description on the dataset page: https://huggingface.co/datasets/FastDOLz/public-company-federal-compliance.clinical-quad-epro-compliance-diary-fatigue-backfill-endpoint-reliability-loss-v0.1Clinical Quad ePRO Compliance Diary Fatigue Backfill Endpoint Reliability Loss v0.1
Each row is a site monthly snapshot.
Core quad
ePRO complianceDiary fatigueBackfill entriesEndpoint reliability loss
Target
label_primary_fail_next_90d
Files
data/train.csvdata/tester.csvscorer.py
Evaluation
Run model on data/tester.csvReturn predictions row alignedScore with scorer.py
License
MIT
This dataset identifies a measurable coupling pattern associated with systemic instability.
The sample demonstrates… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/clinical-quad-epro-compliance-diary-fatigue-backfill-endpoint-reliability-loss-v0.1.healthcare-voc-compliance
Healthcare VOC Compliance Datasets
Two structured datasets for evaluating Volatile Organic Compound (VOC) regulatory compliance in healthcare facility cleaning across 26 North American jurisdictions.
Intended Use
Compliance modeling — determine whether a cleaning product meets VOC limits in a given jurisdiction before procurement
Product selection — filter and rank healthcare cleaning products by VOC content, certification status, and multi-jurisdiction compliance… See the full description on the dataset page: https://huggingface.co/datasets/davecook1985/healthcare-voc-compliance.Investing-Compliancelegal-undertaking-order-obligation-compliance-coherence-risk-v0.1What this dataset does
You receive
obligation text
deadline
owner
required proof
action log
status
You decide
coherent
or
incoherent
Daily use
order compliance tracking
breach risk flagging
missing proof detection
public-companies-federal-compliance_q3
This is a mirror. Canonical home: https://www.fastdol.com/datasets/public-company-federal-compliance
License: CC BY 4.0
Version DOI: 10.5281/zenodo.20613452
Concept DOI: 10.5281/zenodo.20031893
Download CSV: https://www.fastdol.com/datasets/public-company-federal-compliance/data.csv
Visit the canonical page for the full schema, methodology, BibTeX citation, and most recent version.
Public Company Federal Compliance Records — Q3 2026
US workplace enforcement… See the full description on the dataset page: https://huggingface.co/datasets/FastDOL/public-companies-federal-compliance_q3.tax-compliance-us
US Tax Compliance Returns Dataset
Tax return filings for a simulated US retail SME over 2 financial years. Each return pulls real figures from the general ledger. Useful for tax software testing, compliance verification, and accounting education.
Tables
Table
Rows
companies
1
tax_return_lines
56
tax_returns
8
Total
65
Formats
CSV (in data/ directory)
Parquet (available in full version)
SQLite (available in full version)
About… See the full description on the dataset page: https://huggingface.co/datasets/mindweave/tax-compliance-us.Insurance-Chatbot-Cross-border-Compliance-Harmless
Dataset Card for Cross-border Compliance Harmless
Description
The test set is designed to assess the performance of chatbots in the telecom and insurance industries, specifically focusing on the behaviors of reliability and the categories of harmless responses. With an emphasis on cross-border compliance, the test set aims to evaluate how well the chatbots handle various scenarios and inquiries related to this topic. Through comprehensive testing, chatbot developers can… See the full description on the dataset page: https://huggingface.co/datasets/rhesis/Insurance-Chatbot-Cross-border-Compliance-Harmless.ai-5node-reg-buf-lag-cpl-compliance-break-v0.1
What this repo does
This dataset models compliance break cascades in AI deployment and market expansion. It detects when regulatory pressure rises, compliance buffers weaken, governance lag delays remediation and evidence capture, and tight coupling through shared templates and cross-product rollouts crosses the five-node cascade threshold into an unrecoverable compliance break cascade.
This dataset models a five-node cascade: four interacting instability drivers and one emergent… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/ai-5node-reg-buf-lag-cpl-compliance-break-v0.1.Investing-Compliancedga-compliance-datasetThis dataset contains 2,000,000 domain name records with 20 handcrafted lexical features extracted from second-level domain strings, constructed for binary classification of Domain Generation Algorithm (DGA) generated domains against legitimate domains. The dataset is perfectly balanced, comprising 1,000,000 DGA domain strings sampled from the ExtraHop golden-rules corpus and 1,000,000 benign domain strings sampled from the Tranco top-one-million popularity ranking list. Features span four… See the full description on the dataset page: https://huggingface.co/datasets/williegeodev/dga-compliance-dataset.ComplianceLLMtreaty_compliance