# ROMAN: Reward-Orchestrated Multi-Head Attention Network for Autonomous Driving System Testing

arXiv: https://arxiv.org/abs/2602.05629 · HTML: https://mlsanity.com/paper/2602.05629/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 38 references that are in the map.
- First posted on arXiv 5 Feb 2026.

## Builds on (most cited first, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation](https://mlsanity.com/paper/2106.04399/) (2021)
- [Self-Driving Cars: A Survey](https://mlsanity.com/paper/1901.04407/) (2019)
- [LGSVL Simulator: A High Fidelity Simulator for Autonomous Driving](https://mlsanity.com/paper/2005.03778/) (2020)
- [Multi-modal Motion Prediction with Transformer-based Neural Network for Autonomous Driving](https://mlsanity.com/paper/2109.06446/) (2021)
- [A Survey on Automated Driving System Testing: Landscapes and Trends](https://mlsanity.com/paper/2206.05961/) (2022)
- [Corner Case Generation and Analysis for Safety Assessment of Autonomous Vehicles](https://mlsanity.com/paper/2102.03483/) (2021)
- [RTAMT: Online Robustness Monitors from STL](https://mlsanity.com/paper/2005.11827/) (2020)
- [Testing of Autonomous Driving Systems: Where Are We and Where Should We Go?](https://mlsanity.com/paper/2106.12233/) (2021)
- [LawBreaker: An Approach for Specifying Traffic Laws and Fuzzing Autonomous Vehicles](https://mlsanity.com/paper/2208.14656/) (2022)
- [BehAVExplor: Behavior Diversity Guided Testing for Autonomous Driving Systems](https://mlsanity.com/paper/2307.07493/) (2023)
- [Generating Out-Of-Distribution Scenarios Using Language Models](https://mlsanity.com/paper/2411.16554/) (2024)
- [Clustering-based Criticality Analysis for Testing of Automated Driving Systems](https://mlsanity.com/paper/2306.12738/) (2023)
- [Dance of the ADS: Orchestrating Failures through Historically-Informed Scenario Fuzzing](https://mlsanity.com/paper/2407.04359/) (2024)

## Abstract (arXiv)

Automated Driving System (ADS) acts as the brain of autonomous vehicles, responsible for their safety and efficiency. Safe deployment requires thorough testing in diverse real-world scenarios and compliance with traffic laws like speed limits, signal obedience, and right-of-way rules. Violations like running red lights or speeding pose severe safety risks. However, current testing approaches face significant challenges: limited ability to generate complex and high-risk law-breaking scenarios, and failing to account for complex interactions involving multiple vehicles and critical situations. To address these challenges, we propose ROMAN, a novel scenario generation approach for ADS testing that combines a multi-head attention network with a traffic law weighting mechanism. ROMAN is designed to generate high-risk violation scenarios to enable more thorough and targeted ADS evaluation. The multi-head attention mechanism models interactions among vehicles, traffic signals, and other factors. The traffic law weighting mechanism implements a workflow that leverages an LLM-based risk weighting module to evaluate violations based on the two dimensions of severity and occurrence. We have evaluated ROMAN by testing the Baidu Apollo ADS within the CARLA simulation platform and conducting extensive experiments to measure its performance. Experimental results demonstrate that ROMAN surpassed state-of-the-art tools ABLE and LawBreaker by achieving 7.91% higher average violation count than ABLE and 55.96% higher than LawBreaker, while also maintaining greater scenario diversity. In addition, only ROMAN successfully generated violation scenarios for every clause of the input traffic laws, enabling it to identify more high-risk violations than existing approaches.