# Realistic Extreme Behavior Generation for Improved AV Testing

arXiv: https://arxiv.org/abs/2409.10669 · HTML: https://mlsanity.com/paper/2409.10669/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Sep 2024.

## Built on by (most cited first, 3 of 3)

- [Causal Composition Diffusion Model for Closed-loop Traffic Generation](https://mlsanity.com/paper/2412.17920/) (2024)
- [Model-Based Policy Adaptation for Closed-Loop End-to-End Autonomous Driving](https://mlsanity.com/paper/2511.21584/) (2025)
- [Realistic adversarial scenario generation via human-like pedestrian model for autonomous vehicle control parameter optimisation](https://mlsanity.com/paper/2601.02082/) (2026)

## Abstract (arXiv)

This work introduces a framework to diagnose the strengths and shortcomings of Autonomous Vehicle (AV) collision avoidance technology with synthetic yet realistic potential collision scenarios adapted from real-world, collision-free data. Our framework generates counterfactual collisions with diverse crash properties, e.g., crash angle and velocity, between an adversary and a target vehicle by adding perturbations to the adversary's predicted trajectory from a learned AV behavior model. Our main contribution is to ground these adversarial perturbations in realistic behavior as defined through the lens of data-alignment in the behavior model's parameter space. Then, we cluster these synthetic counterfactuals to identify plausible and representative collision scenarios to form the basis of a test suite for downstream AV system evaluation. We demonstrate our framework using two state-of-the-art behavior prediction models as sources of realistic adversarial perturbations, and show that our scenario clustering evokes interpretable failure modes from a baseline AV policy under evaluation.