Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure
Summary
This paper diagnoses long-horizon failures in world models, attributing them to kinematic rather than dynamic imagination. The authors introduce a metric (iKCE) and show that imagined rollouts fail to capture dynamic regime changes even as policy rewards collapse.
View Cached Full Text
Cached at: 07/09/26, 03:40 PM
Paper page - Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure
Source: https://huggingface.co/papers/2607.05966
Abstract
World models exhibit long-horizon failures due to kinematic rather than dynamic imagination, as demonstrated by measuring imagined kinematic-consistency error which remains flat while policy rewards collapse across friction boundaries.
Long-horizon failure inworld modelsis conventionally attributed tocompounding error, a generic framing that does not distinguish what kind of error compounds. We propose akinematic-vs-dynamic reframing:world modelstend to imagine kinematically rather than dynamically. We operationalize this as the imaginedKinematic-Consistency Error, a per-step diagnostic that measures how far arolloutdeparts from aclosed-form kinematic null, paired with aperturbation protocolthat tests whetheriKCEresponds when physical conditions cross a regime boundary. We instantiate the diagnostic on a releasedDreamerV3checkpoint trained onDMC walker-walk, where imaginediKCEruns roughly two orders of magnitude above that of matched real-physicsrollouts. Across a friction sweep that crosses thegait-collapse boundary, the model’siKCEstays statistically flat even as the trained policy’s reward collapses through the same range, providing the kinematic-not-dynamic signature. The diagnostic distinguishes kinematic from dynamic imagination at horizons longer than theembodiment’s gait period.
View arXiv pageView PDFGitHub1Add to collection
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2607.05966 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2607.05966 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2607.05966 in a Space README.md to link it from this page.
Collections including this paper1
Similar Articles
Understanding Rollout Error in Graph World Models
This paper analyzes long-horizon rollout error in Graph World Models (GWMs), proposing a unified framework with dynamic edges and introducing Error-Aware GWM that uses spectral regularization, rollout consistency, and critical-node weighting to prevent divergence.
Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors [R]
This paper introduces a bidirectional latent diffusion model that steps dynamical systems forward or backward in time, using round-trip consistency as a self-supervised test-time error signal to predict rollout errors without ground truth or ensembles.
Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform
This paper argues that large language models struggle with causal reasoning and long-horizon planning due to a mismatch between sequence prediction and reasoning over latent environment dynamics, and introduces the Latent Dynamics Inference perspective along with the Flux environment to study these limitations.
GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation
This paper systematically studies world models for robotic policy evaluation, introduces the WMBench benchmark and GigaWorld-1 model, and shows that long-horizon rollout consistency is more critical than short-term visual realism.
The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World
This paper demonstrates that replay-based static evaluation of model switching in LLM agents is fundamentally flawed: when swapping models mid-trajectory, the environment and subsequent actions diverge dramatically from logged trajectories, invalidating most benchmark results. The authors propose branching rollouts as a more faithful evaluation method and release their harness and trajectories.