# Validate the Dream Before You Trust Its Verdict: Admissibility for World-Model Simulators arXiv: https://arxiv.org/abs/2607.07196 ยท HTML: https://mlsanity.com/paper/2607.07196/ ## Key facts - Cited by 2 papers in the map (data through 14 Oct 2026). - Placed in Embodied and robot world models, one of 11 directions in World models. - Builds on 30 of its 44 references that are in the map. - First posted on arXiv 8 Jul 2026. ## Placed in - World models / [Embodied and robot world models](https://mlsanity.com/wm/embodied-and-robot-world-models/) ## Builds on (most cited first, 30 of 30) - [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023) - [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016) - [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018) - [Towards Accurate Generative Models of Video: A New Metric & Challenges](https://mlsanity.com/paper/1812.01717/) (2018) - [A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification](https://mlsanity.com/paper/2107.07511/) (2021) - [Recurrent World Models Facilitate Policy Evolution](https://mlsanity.com/paper/1809.01999/) (2018) - [Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models](https://mlsanity.com/paper/1805.12114/) (2018) - [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023) - [GAIA-1: A Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2309.17080/) (2023) - [VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://mlsanity.com/paper/2303.16727/) (2023) - [Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability](https://mlsanity.com/paper/2405.17398/) (2024) - [Pseudo-Simulation for Autonomous Driving](https://mlsanity.com/paper/2506.04218/) (2025) - [Epona: Autoregressive Diffusion World Model for Autonomous Driving](https://mlsanity.com/paper/2506.24113/) (2025) - [DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving](https://mlsanity.com/paper/2408.00415/) (2024) - [WorldGym: World Model as An Environment for Policy Evaluation](https://mlsanity.com/paper/2506.00613/) (2025) - [Sample-Efficient Safety Assurances using Conformal Prediction](https://mlsanity.com/paper/2109.14082/) (2021) - [Do World Action Models Generalize Better than VLAs? A Robustness Study](https://mlsanity.com/paper/2603.22078/) (2026) - [The Reality Gap in Robotics: Challenges, Solutions, and Best Practices](https://mlsanity.com/paper/2510.20808/) (2025) - [Beyond FVD: Enhanced Evaluation Metrics for Video Generation Quality](https://mlsanity.com/paper/2410.05203/) (2024) - [DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving](https://mlsanity.com/paper/2601.01528/) (2026) - [Are NeRFs ready for autonomous driving? Towards closing the real-to-simulation gap](https://mlsanity.com/paper/2403.16092/) (2024) - [World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry](https://mlsanity.com/paper/2604.01985/) (2026) - [Do Open-Loop Metrics Predict Closed-Loop Driving? A Cross-Benchmark Correlation Study of NAVSIM and Bench2Drive](https://mlsanity.com/paper/2605.00066/) (2026) - [From Generative Engines to Actionable Simulators: The Imperative of Physical Grounding in World Models](https://mlsanity.com/paper/2601.15533/) (2026) - [Black-Box Safety Validation of Autonomous Systems: A Multi-Fidelity Reinforcement Learning Approach](https://mlsanity.com/paper/2203.03451/) (2022) - [Beyond Simulation: Benchmarking World Models for Planning and Causality in Autonomous Driving](https://mlsanity.com/paper/2508.01922/) (2025) - [Latent World Models for Automated Driving: A Unified Taxonomy, Evaluation Framework, and Open Challenges](https://mlsanity.com/paper/2603.09086/) (2026) - [Target-Bench: Can Video World Models Achieve Mapless Path Planning with Semantic Targets?](https://mlsanity.com/paper/2511.17792/) (2025) - [Analyzing the Impact of Simulation Fidelity on the Evaluation of Autonomous Driving Motion Control](https://mlsanity.com/paper/2602.07984/) (2026) - [Disengagement Analysis and Field Tests of a Prototypical Open-Source Level 4 Autonomous Driving System](https://mlsanity.com/paper/2603.21926/) (2026) ## Built on by (most cited first, 2 of 2) - [DreamLedger: Where to Refuse World-Model Imagination Using Execution-Settled Credit](https://mlsanity.com/paper/2608.23863/) (2026) - [SimVerity: When Does Simulated Agent Success Survive Physical Deployment?](https://mlsanity.com/paper/2608.25067/) (2026) ## Abstract (arXiv) Across robotics, World Models (WMs) are increasingly used to evaluate action policies by simulating the consequences of actions in an imagined world, and returning a success or safety verdict. Yet a verdict is only as trustworthy as the WM that produced it, and the WM itself needs to be certified. In video-generation WMs, fidelity metrics such as Fr'echet Video Distance (FVD) reward visual realism, but ignore whether the world responds correctly to the policy's actions, including those unseen in training. Classical simulation-based validation assumes a trusted simulator evaluating an untrusted policy, whereas generative WMs are themselves unverified learned artifacts. Hence, we argue that any WM used as a test oracle must first be accredited before its verdicts can serve as evidence. Building on credibility practices from safety-critical simulation, including Verification, Validation & Accreditation (VV&A), Safety of the Intended Functionality (SOTIF), and scenario-based testing standards, we define an admissibility ladder (L0-L4) that a WM must climb before its closed-loop verdicts are accepted as assurance evidence. Our framework is embodiment-agnostic, and is instantiated in autonomous driving (AD), where assurance methods for traditional simulation are most mature. Applied to two driving WMs, the lower rungs reveal a reversal: the model that ranks higher on visual generation quality (L0) ranks lower on action-following (L1-L2), so visual fidelity does not predict the action-robustness a closed-loop verdict depends on.