Multi-agent behavior modeling and trajectory forecasting are crucial for the safe navigation of autonomous agents in interactive scenarios. Variational Autoencoder (VAE) has been widely applied in multi-agent interaction modeling to generate diverse behavior and learn a low-dimensional representation for interacting systems. However, existing literature did not formally discuss if a VAE-based model can properly encode interaction into its latent space. In this work, we argue that one of the typical formulations of VAEs in multi-agent modeling suffers from an issue we refer to as social posterior collapse, i.e., the model is prone to ignoring historical social context when predicting the future trajectory of an agent. It could cause significant prediction errors and poor generalization performance.

In this section, we report the testing results on Argoverse and ETH/UCY datasets and compare the results with other models in the literature.

Also, the proposed measures are effective in alleviating the issue.

In this section, we report the testing results on Argoverse and ETH/UCY datasets and compare the results with other models in the literature.

Also, the proposed measures are effective in alleviating the issue.