Paper 2604.28196
HERMES++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation
- Published
- Apr 2026
- Research lab
- Independent
- Citations
- 3
- GitHub
- 69 stars
01 In brief
Summary
HERMES++ is a unified driving world model that integrates 3D scene understanding and future geometry prediction within a single framework.
It uses a Bird's-Eye View (BEV) representation to consolidate multi-view spatial information, making it compatible with Large Language Models (LLMs).
The model introduces LLM-enhanced world queries for knowledge transfer and a Current-to-Future Link to bridge the temporal gap, conditioning geometric evolution on semantic context.
A Joint Geometric Optimization strategy enforces structural integrity by combining explicit geometric constraints with implicit latent regularization.
Evaluations on multiple benchmarks show HERMES++ outperforms specialist approaches in both future point cloud prediction and 3D scene understanding, achieving a 8.2% error reduction over DriveX for 3s point cloud generation and a 9.2% improvement over Omni-Q on the OmniDrive-nuScenes CIDEr metric.
The model also demonstrates strong generalization to NuScenes-QA, DriveLM, and motion planning tasks.
02 From the paper
Abstract
Driving world models serve as a pivotal technology for autonomous driving by simulating environmental dynamics. However, existing approaches predominantly focus on future scene generation, often overlooking comprehensive 3D scene understanding. Conversely, while Large Language Models (LLMs) demonstrate impressive reasoning capabilities, they lack the capacity to predict future geometric evolution, creating a significant disparity between semantic interpretation and physical simulation. To bridge this gap, we propose HERMES++, a unified driving world model that integrates 3D scene understanding and future geometry prediction within a single framework. Our approach addresses the distinct requirements of these tasks through synergistic designs. First, a BEV representation consolidates multi-view spatial information into a structure compatible with LLMs. Second, we introduce LLM-enhanced world queries to facilitate knowledge transfer from the understanding branch. Third, a Current-to-Future Link is designed to bridge the temporal gap, conditioning geometric evolution on semantic context. Finally, to enforce structural integrity, we employ a Joint Geometric Optimization strategy that integrates explicit geometric constraints with implicit latent regularization to align internal representations with geometry-aware priors. Extensive evaluations on multiple benchmarks validate the effectiveness of our method. HERMES++ achieves strong performance, outperforming specialist approaches in both future point cloud prediction and 3D scene understanding tasks. The model and code will be publicly released at https://github.com/H-EmbodVis/HERMESV2.