
This preprint introduces VERIDEX V9.1, a Policy-Latent Diffusion Network (PLD-Net) designed for multi-country content rating prediction across 65 countries and 51 rating classes. The model achieves 80.6% validation accuracy and 80.3% test accuracy on a dataset of 12,264 movies, improving baseline performance by +3.48%. VERIDEX V9.1 introduces four novel contributions:(1) Uncertainty-Weighted Policy Ensemble (UWPE),(2) Hierarchical Multi-Head Policy Attention (HMPA),(3) Policy Consistency Regularization (PCR),(4) Progressive Knowledge Distillation (PKD). This version includes complete reproducibility details, architecture diagrams, dataset specifications, TMDb-compliance notes, ablation studies, and evaluation metrics. The work is released as a research preprint for academic visibility and citation.
AI safety & compliance, machine learning, transformer models, policy learning, latent diffusion, AI safety & compliance, content rating prediction, multi-country analysis, deep learning, uncertainty ensemble
AI safety & compliance, machine learning, transformer models, policy learning, latent diffusion, AI safety & compliance, content rating prediction, multi-country analysis, deep learning, uncertainty ensemble
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
