DrivingDojo Dataset: Advancing Interactive and Knowledge-Enriched Driving World Model

NeurIPS 24 2024 61.9 benchmark, application

TLDR

DrivingDojo dataset enables interactive world models with diverse driving maneuvers and action-controlled future prediction benchmark.

Reasoning

The paper introduces a novel dataset addressing video diversity limitations for interactive driving world models, with strengths in maneuver diversity and benchmark definition. Weaknesses include lack of explicit real-world validation details and limited results in the abstract.

Read-first score

Read-first score 61.9, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 49.

Recency 8%

75.1

Uses a gentle age decay so recent papers surface without erasing older foundations. 2024

Topical relevance 42%

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

Methodology quality 25%

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

Reproducibility 25%

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

Field roles

Candidate

Rank sensitivity

Stability: volatile; rank range: 192.

Keyword Scores

interactive world model

world model

video world model

world dynamics prediction

generative world model

world simulator

model-based reinforcement learning world model

Deep Analysis

Innovations

First dataset tailor-made for training interactive world models with complex driving dynamics
Includes video clips with a complete set of driving maneuvers, diverse multi-agent interplay, and rich open-world driving knowledge
Defines an action instruction following (AIF) benchmark for evaluating world models on action-controlled future predictions

Methodology

The paper introduces DrivingDojo, a dataset featuring video clips with complete driving maneuvers, multi-agent interplay, and open-world driving knowledge. It also defines an Action Instruction Following (AIF) benchmark to evaluate world models on action-controlled future prediction tasks.

Key Results

The proposed dataset demonstrates superiority for generating action-controlled future predictions compared to existing datasets.

Resources

Paper Project