Same World, Different Knowledge: When Isolated Audits Misjudge World-Model Repairs
New in World models; no direction yet. World models
Abstract (arXiv)
A repair favored under an isolated input fault can be inferior when deployed modules share the faulty information. We introduce an information-interface audit for world models, distinguishing fidelity gaps, where exact inputs become estimates, from availability gaps, where inputs are missing. Fixed-weight interventions measure prediction error, input dependence, and paired closed-loop benefit, including dependencies introduced by reconstruction. In simulated quadrotor model predictive control, coupled, opposite-sign 10% mass/thrust calibration errors reduce a physics-anchored model's success from 69% to 8%; uncertainty training restores 65%. Wind reconstruction recovers control benefit but inherits calibration dependence. For a positive calibration offset, reconstruction-only corruption favors uncertainty-trained reconstruction, whereas shared corruption favors the baseline. Acceleration diagnostics reveal compensation between reconstruction bias and nominal-model error, also observed with a disturbance observer. Repair selection therefore depends on the information paths used in deployment.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 6 of its 27 references that are in the map. First posted on arXiv 17 Sep 2026.
Builds on
6 of 27 references.
References · 6
- 2023A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learningcited by 9
- 2026SkyJEPA: Learning Long-Horizon World Models for Zero-Shot Sim-to-Real Control of Quadrotorscited by 9
- 2018Adaptive Model Predictive Control for High-Accuracy Trajectory Tracking in Changing Conditionscited by 7
- 2026Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planningcited by 7
- 2026Predicting Closed-Loop Performance of Latent World Models: Offline Checkpoint Selection for MPC and Model-Based RL Under Non-Markovian Rewards in LunarLandercited by 2
- 2024Fixed-time Disturbance Observer-Based MPC Robust Trajectory Tracking Control of Quadrotorcited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Learning from World Feedback: Why Model Uncertainty Fails as a Risk Signal in Model-Based RL
- 2026World Models Dream of Success: Diagnosing and Repairing Failure Insensitivity in Robot World Models
- 2026MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination
- 2025World Models for Anomaly Detection during Model-Based Reinforcement Learning Inference
- 2026When World Models Lie: Adaptive Safety Analysis Under Wrong Imaginations
- 2026Evaluation Blindness: How Silent Measurement Failures Corrupt AI Systems from Training to Deployment
- 2026Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes
- 2026REFLEX: A Learned On-Board Reflex for Robust Autonomous Collision Recovery of Soft Quadrotors ↗