MDG: Masked Denoising Generation for Multi-Agent Behavior Modeling in Traffic Environments

TL;DR

MDG introduces a unified, efficient framework for multi-agent behavior modeling in traffic environments, enabling controllable and versatile trajectory generation without iterative sampling or task-specific designs.

Contribution

The paper presents MDG, a novel masked denoising generative approach that reformulates multi-agent behavior modeling as tensor reconstruction, improving efficiency and versatility over existing methods.

Findings

Achieves competitive performance on Waymo Sim Agents benchmark.

Provides efficient open-loop multi-agent trajectory generation.

Unifies various behavior modeling tasks within a single framework.

Abstract

Modeling realistic and interactive multi-agent behavior is critical to autonomous driving and traffic simulation. However, existing diffusion and autoregressive approaches are limited by iterative sampling, sequential decoding, or task-specific designs, which hinder efficiency and reuse. We propose Masked Denoising Generation (MDG), a unified generative framework that reformulates multi-agent behavior modeling as the reconstruction of independently noised spatiotemporal tensors. Instead of relying on diffusion time steps or discrete tokenization, MDG applies continuous, per-agent and per-timestep noise masks that enable localized denoising and controllable trajectory generation in a single or few forward passes. This mask-driven formulation generalizes across open-loop prediction, closed-loop simulation, motion planning, and conditional generation within one model. Trained on large-scale real-world driving datasets, MDG achieves competitive closed-loop performance on the Waymo Sim Agents and nuPlan Planning benchmarks, while providing efficient, consistent, and controllable open-loop multi-agent trajectory generation. These results position MDG as a simple yet versatile paradigm for multi-agent behavior modeling.