Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Flow-ERD, a multi-agent traffic simulator that effectively balances realism and diversity in autonomous driving scenarios. By employing Agent-Type Aware Flow Matching (AFM) for kinematic execution tailored to agent types and utilizing Entropy-Regularized Distillation (ERD) to refine the distribution of simulated traffic, the method addresses the shortcomings of existing benchmarks that prioritize realism at the expense of diversity. Flow-ERD achieves superior performance on the WOSAC test benchmark, ranking first and demonstrating a significant improvement in the realism-diversity trade-off compared to existing methods.
Flow-ERD not only excels in realism but also preserves fine-grained diversity, outperforming existing benchmarks in traffic simulation.
Realistic and diverse traffic simulation is essential to autonomous driving development. Yet prevailing benchmarks predominantly reward realism, and recent methods have optimized accordingly, leaving diversity underexplored. We introduce \textbf{Flow-ERD}, a multi-agent simulator that pursues realism and diversity jointly. Its backbone, \textbf{Agent-Type Aware Flow Matching} (AFM), couples flow matching's multi-modal expressiveness with type-specific kinematic execution. It preserves fine-grained diversity while keeping motions consistent with each agent type. A second stage, \textbf{Entropy-Regularized Distillation} (ERD), fine-tunes the closed-loop rollout distribution with an entropy-regularized reverse-KL objective. This mitigates covariate shift while explicitly preventing collapse onto high-density modes. We evaluate Flow-ERD with a log-free diversity metric alongside standard realism scores. Flow-ERD ranks first on the WOSAC test benchmark and dominates the realism--diversity Pareto front among reproducible baselines. Our project page is available \href{https://seulbinhwang.github.io/flow-erd-project-page/}{here}.