# DecompGAIL: Learning Realistic Traffic Behaviors with Decomposed Multi-Agent Generative Adversarial Imitation Learning

arXiv: https://arxiv.org/abs/2510.06913 · HTML: https://mlsanity.com/paper/2510.06913/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Imitation and offline RL for driving, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 35 of its 60 references that are in the map.
- First posted on arXiv 8 Oct 2025.

## Placed in

- Offline RL and imitation learning from data / [Imitation and offline RL for driving](https://mlsanity.com/offlinerl/imitation-and-offline-rl-for-driving/)

## Builds on (most cited first, 30 of 35)

- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [Which Training Methods for GANs do actually Converge?](https://mlsanity.com/paper/1801.04406/) (2018)
- [Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Dataset](https://mlsanity.com/paper/2104.10133/) (2021)
- [Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge?](https://mlsanity.com/paper/2011.09533/) (2020)
- [A Divergence Minimization Perspective on Imitation Learning Methods](https://mlsanity.com/paper/1911.02256/) (2019)
- [Multi-Agent Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1807.09936/) (2018)
- [Imitating Driver Behavior with Generative Adversarial Networks](https://mlsanity.com/paper/1701.06699/) (2017)
- [The Waymo Open Sim Agents Challenge](https://mlsanity.com/paper/2305.12032/) (2023)
- [BITS: Bi-level Imitation for Traffic Simulation](https://mlsanity.com/paper/2208.12403/) (2022)
- [SimNet: Learning Reactive Self-driving Simulations from Real-world Observations](https://mlsanity.com/paper/2105.12332/) (2021)
- [GoRela: Go Relative for Viewpoint-Invariant Motion Forecasting](https://mlsanity.com/paper/2211.02545/) (2022)
- [Multi-Agent Adversarial Inverse Reinforcement Learning](https://mlsanity.com/paper/1907.13220/) (2019)
- [Trajeglish: Traffic Modeling as Next-Token Prediction](https://mlsanity.com/paper/2312.04535/) (2023)
- [Multi-Agent Imitation Learning for Driving Simulation](https://mlsanity.com/paper/1803.01044/) (2018)
- [BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction](https://mlsanity.com/paper/2405.17372/) (2024)
- [Modeling Human Driving Behavior through Generative Adversarial Imitation Learning](https://mlsanity.com/paper/2006.06412/) (2020)
- [Solving Motion Planning Tasks with a Scalable Generative Model](https://mlsanity.com/paper/2407.02797/) (2024)
- [Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Models](https://mlsanity.com/paper/2412.05334/) (2024)
- [PredictionNet: Real-Time Joint Probabilistic Traffic Prediction for Planning, Control, and Simulation](https://mlsanity.com/paper/2109.11094/) (2021)
- [Improving Agent Behaviors with RL Fine-tuning for Autonomous Driving](https://mlsanity.com/paper/2409.18343/) (2024)
- [Learning from Demonstration in the Wild](https://mlsanity.com/paper/1811.03516/) (2018)
- [SMART: Scalable Multi-agent Real-time Motion Generation via Next-token Prediction](https://mlsanity.com/paper/2405.15677/) (2024)
- [Simulating Emergent Properties of Human Driving Behavior Using Multi-Agent Reward Augmented Imitation Learning](https://mlsanity.com/paper/1903.05766/) (2019)
- [KiGRAS: Kinematic-Driven Generative Model for Realistic Agent Simulation](https://mlsanity.com/paper/2407.12940/) (2024)
- [Advancing Multi-agent Traffic Simulation via R1-Style Reinforcement Fine-Tuning](https://mlsanity.com/paper/2509.23993/) (2025)
- [UniMM: A Unified Mixture Model Framework for Multi-Agent Simulation](https://mlsanity.com/paper/2501.17015/) (2025)
- [CCIL: Context-conditioned imitation learning for urban driving](https://mlsanity.com/paper/2305.02649/) (2023)
- [LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation](https://mlsanity.com/paper/2403.17601/) (2024)
- [TrajTok: Technical Report for 2025 Waymo Open Sim Agents Challenge](https://mlsanity.com/paper/2506.21618/) (2025)
- [Learning to Simulate on Sparse Trajectory Data](https://mlsanity.com/paper/2103.11845/) (2021)

## Built on by (most cited first, 5 of 5)

- [AutoWorld: Learning Multi-Agent Traffic Simulation with Self-Supervised World Models](https://mlsanity.com/paper/2603.28963/) (2026)
- [Long-term Traffic Simulation via Structured Autoregressive Modeling](https://mlsanity.com/paper/2606.31209/) (2026)
- [General-purpose LLMs as Models of Human Driver Behavior: The Case of Simplified Merging](https://mlsanity.com/paper/2604.09609/) (2026)
- [Artificial Intelligence for Modeling and Simulation of Mixed Automated and Human Traffic](https://mlsanity.com/paper/2604.12857/) (2026)
- [Flow-ERD: Agent-type Aware Flow Matching with Entropy-Regularized Distillation for Diverse Traffic Simulation](https://mlsanity.com/paper/2607.06957/) (2026)

## Abstract (arXiv)

Realistic traffic simulation is critical for the development of autonomous driving systems and urban mobility planning, yet existing imitation learning approaches often fail to model realistic traffic behaviors. Behavior cloning suffers from covariate shift, while Generative Adversarial Imitation Learning (GAIL) is notoriously unstable in multi-agent settings. We identify a key source of this instability: irrelevant interaction misguidance, where a discriminator penalizes an ego vehicle's realistic behavior due to unrealistic interactions among its neighbors. To address this, we propose Decomposed Multi-agent GAIL (DecompGAIL), which explicitly decomposes realism into ego-map and ego-neighbor components, filtering out misleading neighbor: neighbor and neighbor: map interactions. We further introduce a social PPO objective that augments ego rewards with distance-weighted neighborhood rewards, encouraging overall realism across agents. Integrated into a lightweight SMART-based backbone, DecompGAIL achieves state-of-the-art performance on the WOMD Sim Agents 2025 benchmark.