# Evaluating Physical Consistency and Plausibility in Generative Scenario Models for Autonomous Driving

arXiv: https://arxiv.org/abs/2610.01581 · HTML: https://mlsanity.com/paper/2610.01581/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 40 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Builds on (most cited first, 18 of 18)

- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [An Introduction to Variational Autoencoders](https://mlsanity.com/paper/1906.02691/) (2019)
- [Understanding disentangling in β-VAE](https://mlsanity.com/paper/1804.03599/) (2018)
- [AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning](https://mlsanity.com/paper/2506.13757/) (2025)
- [Guided Conditional Diffusion for Controllable Traffic Simulation](https://mlsanity.com/paper/2210.17366/) (2022)
- [GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2503.20523/) (2025)
- [A Survey on Safety-Critical Driving Scenario Generation -- A Methodological Perspective](https://mlsanity.com/paper/2202.02215/) (2022)
- [Explainable Artificial Intelligence for Autonomous Driving: A Comprehensive Overview and Field Guide for Future Research Directions](https://mlsanity.com/paper/2112.11561/) (2021)
- [A Survey for Foundation Models in Autonomous Driving](https://mlsanity.com/paper/2402.01105/) (2024)
- [Scenarios for Development, Test and Validation of Automated Vehicles](https://mlsanity.com/paper/1801.08598/) (2018)
- [Interpretable Self-Aware Neural Networks for Robust Trajectory Prediction](https://mlsanity.com/paper/2211.08701/) (2022)
- [Variational Autoencoder-Based Vehicle Trajectory Prediction with an Interpretable Latent Space](https://mlsanity.com/paper/2103.13726/) (2021)
- [A Survey on Scenario-Based Testing for Automated Driving Systems in High-Fidelity Simulation](https://mlsanity.com/paper/2112.00964/) (2021)
- [Realistic Corner Case Generation for Autonomous Vehicles with Multimodal Large Language Model](https://mlsanity.com/paper/2412.00243/) (2024)
- [Beta-VAE Reproducibility: Challenges and Extensions](https://mlsanity.com/paper/2112.14278/) (2021)
- [Enhancing Physical Consistency in Lightweight World Models](https://mlsanity.com/paper/2509.12437/) (2025)
- [Scenario Parameter Generation Method and Scenario Representativeness Metric for Scenario-Based Assessment of Automated Vehicles](https://mlsanity.com/paper/2202.12025/) (2022)

## Abstract (arXiv)

Generative AI models are increasingly used for scenario generation in autonomous driving. While they can generate realistic-looking scenarios, they often provide limited transparency into learned representations and consistency with real-world vehicle dynamics. This lack of formal assurance limits their use in safety-critical validation and certification workflows. To address this aspect, we introduce a layered evaluation protocol that complements existing methods by assessing models across five layers. The first four layers inspect internal representations and network layers through kinematic alignment, statistical baseline comparison, latent controllability, and activation analysis. The fifth layer evaluates model outputs against vehicle dynamics constraints such as lateral jerk thresholds. We demonstrate the protocol on a Variational Autoencoder (VAE)-based scenario generator. Although standard output-level metrics and visualizations suggest that the generated scenarios are realistic, our protocol provides deeper insight into the extent to which the model's latent space aligns with kinematic features and whether visually plausible trajectories satisfy vehicle-dynamics constraints. We further apply the protocol to additional generative models, demonstrating its applicability beyond the VAE architecture.