Awesome World Model Hub Papers · Datasets · Projects
← Back to papers

JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning

arXiv 2026 55.1 method

TLDR

JEDI combines JEPA-style predictive learning with diffusion denoising for an end-to-end latent diffusion world model, achieving competitive results on Atari100k.

Reasoning

The paper introduces a novel integration of JEPA and diffusion for online MBRL, with theoretical motivation and empirical results on Atari100k. Strengths include end-to-end learning and efficiency gains, but the evaluation is limited to a single simulated benchmark without real-world experiments.

Read-first score

Read-first score 55.1, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 51.

Recency 6%
100

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

Citation impact 18%
77.4

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.77414367

Topical relevance 29%
72.9

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

Methodology quality 18%
50

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

Reproducibility 18%
30

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Citation velocity 12%
0

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

Field roles

FoundationFrontierBridge

Rank sensitivity

Stability: volatile; rank range: 432.

Keyword Scores

world model
10
model-based reinforcement learning world model
10
world dynamics prediction
9
generative world model
8
world simulator
6
interactive world model
5
video world model
3

Deep Analysis

Innovations

  • First online end-to-end latent diffusion world model for model-based reinforcement learning
  • Learns latent space directly from diffusion denoising loss within a JEPA framework
  • Uses denoising to learn and predict future latents instead of reconstruction and pretrained models
  • Theoretical motivation linking conventional JEPA objectives to a predictive information bottleneck and conditional diffusion denoising to a predictive-compression decomposition

Methodology

JEDI is a latent diffusion world model for online model-based reinforcement learning. It learns its latent space directly from the diffusion denoising loss using a JEPA framework, where denoising is used to learn and predict future latents rather than relying on reconstruction and pretrained models. The model is trained end-to-end.

Key Results

JEDI is competitive on Atari100k and outperforms the baseline with separately trained latents where directly comparable. Relative to the pixel diffusion baseline, JEDI uses 43% less VRAM, over 3× faster world-model sampling, and 2.5× faster training.

Tags

diffusion world modelsmodel-based reinforcement learningjoint embeddingpredictive representation learningJEPALG