datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
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.CoT-ComplianceBench
CoT-ComplianceBench
CoT-ComplianceBench is the first benchmark that directly audits the chain-of-thought (CoT) reasoning traces of large language models under adversarial prompting — independently of their final outputs.
We introduce the Compliance Divergence Rate (CDR): the fraction of adversarial prompts where a model reasons unsafely in its CoT yet produces a safe-looking final response, creating a blind spot undetectable by output-only evaluators. We also introduce the… See the full description on the dataset page: https://huggingface.co/datasets/BharathGaddam/CoT-ComplianceBench.