policy-shift
PolicyShiftBench
PolicyShiftBench Public Data
Project Page | Paper (arXiv:2607.05910) | Code
This directory contains the cleaned public PolicyShiftBench data package for anonymous review. All splits are released as Parquet files with image bytes embedded in the image column.
Content Warning
PolicyShiftBench is a safety-evaluation dataset. The training and evaluation data intentionally include policy-triggering content across the covered risk categories, including sensitive visual… See the full description on the dataset page: https://huggingface.co/datasets/PolicyShiftGuard/PolicyShiftBench.adaptive-policy-v2.8-withreason
PolicyShiftBench
Project Page | Paper | Code
PolicyShiftBench is a comprehensive benchmark designed to evaluate policy-adaptive image guardrailing. Instead of treating safety as an intrinsic property of an image, PolicyShiftBench tests whether a model can decide whether an image violates the currently supplied policy and generalize to held-out policy definitions.
The benchmark features 2,000 policy-discriminative instances over 265 images, where each image is paired with… See the full description on the dataset page: https://huggingface.co/datasets/PolicyShiftGuard/adaptive-policy-v2.8-withreason.PolicyShiftBench
PolicyShiftBench Public Data
This directory contains the cleaned public PolicyShiftBench data package for anonymous review. All splits are released as Parquet files with image bytes embedded in the image column.
Content Warning
PolicyShiftBench is a safety-evaluation dataset. The training and evaluation data intentionally include policy-triggering content across the covered risk categories, including sensitive visual concepts and potentially harmful text. The data should… See the full description on the dataset page: https://huggingface.co/datasets/PolicyShiftBench/PolicyShiftBench.policy_shift_dataset
Dataset Card for "policy_shift_dataset"
More Information needed