HERMES++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation

arXiv 2026 63 method, application

TLDR

HERMES++ unifies 3D scene understanding and future geometry prediction in a driving world model using BEV, LLM queries, and geometric optimization.

Reasoning

The paper proposes a novel unified framework that bridges semantic understanding and geometric prediction, with strong empirical results on multiple benchmarks. However, the abstract lacks discussion of limitations and does not address interactive or RL-based aspects, limiting its scope.

Read-first score

Read-first score 63, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 44.

Recency 6%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=True; dataset=False; markers=code,github

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=benchmark,evaluation,metric

Citation impact 18%

67.6

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.67615828

Topical relevance 29%

62.9

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Citation velocity 12%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

FrontierBridgeMethodology anchorReproducibility anchor

Rank sensitivity

Stability: volatile; rank range: 469.

Keyword Scores

world model

generative world model

world dynamics prediction

world simulator

video world model

interactive world model

model-based reinforcement learning world model

Deep Analysis

Innovations

Unified driving world model integrating 3D scene understanding and future geometry prediction within a single framework
BEV representation to consolidate multi-view spatial information into a structure compatible with LLMs
LLM-enhanced world queries to facilitate knowledge transfer from the understanding branch
Current-to-Future Link to bridge temporal gap and condition geometric evolution on semantic context
Joint Geometric Optimization strategy integrating explicit geometric constraints with implicit latent regularization

Methodology

HERMES++ employs a unified framework combining 3D scene understanding and future geometry prediction. It uses a BEV representation to align multi-view spatial data with LLMs, introduces LLM-enhanced world queries for cross-task knowledge transfer, designs a Current-to-Future Link to condition geometric evolution on semantic context, and applies Joint Geometric Optimization with explicit geometric constraints and implicit latent regularization to enforce structural integrity.

Key Results

HERMES++ outperforms specialist approaches in both future point cloud prediction and 3D scene understanding tasks across multiple benchmarks.

Resources

Paper Code