AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Hierarchical World Models with Implicit Dynamics

Gaoyue Zhou, Yvonne Wu, Zichen Cui, Nicolas Ballas, Mahmoud Assran, Lerrel Pinto, Yann LeCun

NeurIPS 2026 · Reinforcement learning › Visual and representation RL · approach: Skill learning and diffusion policies · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Implicit-HWM, a hierarchical world model for embodied agents that plans with implicit dynamics to handle long-horizon tasks.

Problem

Long-horizon planning with world models for embodied agents suffers from cumulative errors and search complexity, and existing hierarchical methods rely on rigid macro-actions or latent skills that yield brittle representations and limited fine-grained control.

Approach

A hierarchical world model using implicit dynamics as an alternative to rigid macro-actions or latent skills, intended to support fine-grained control while enabling test-time adaptation via planning.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add hierarchy (high-level diffusion subgoal model) over single-level WM with MPCSingle-level WM · PushT (random goals)success rate: 0.21 -> 0.77Improved
Triple replanning budget (45 replans) for single-level WM to match inference time (ablation)SL-WM · PushT (random goals)success rate: 0.21 -> 0.28No clear effect
Use diffusion noise as implicit action instead of macro-actions (vs HWM)Hierarchical WM · PushT-Aug (random goals)success rate: 0.16 -> 1.00Improved
Use hierarchy over DINO-WM in vision-based PushT, long horizonDINO-WM low level · Vision PushT (d = 75)success rate: 0.00 -> 0.50Improved
Use Implicit-HWM instead of HWM at short horizonHWM · Vision PushT (d = 25)success rate: 0.96 -> 0.90Hurt
Add non-zero latent noise regularization (alpha > 0) (ablation)Implicit-HWM planner · PushT success ratesuccess rateImproved
Increase number of CEM samples (ablation)Implicit-HWM · PushTsuccess rateImproved
Use very small subgoal radius (Delta = 1, 25 subgoals vs Delta = 5, 5 subgoals) (ablation)Implicit-HWM · PushT-Aug (random goals)success rate: 0.96 -> 0.06Hurt
Increase number of subgoals to 25 at Delta = 5 (ablation)Implicit-HWM · PushT-Aug (random goals)success rate: 0.96 -> 0.38Hurt
Add data augmentation to training data (PushT-Aug)H-Diffuser · PushT (random goals)success rate: 0.01 -> 0.27Mixed
Add replanning at every actionDiffuser · PushTsuccess rateNo clear effect
Hierarchy vs reactive baseline on in-distribution fixed goalImplicit-HWM vs GCBC · PushT (fixed goal)success rate: 1.00 vs 1.00No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy