Safety
Characterizing and Mitigating Distributional Bias in Physics-Informed Generative Models
arXiv:2610.16582v1 Announce Type: new Abstract: Physics-informed generative models augment a data-fitting objective with residual penalties that encourage physical constraints to hold in generated sam
arXiv:2610.16582v1 Announce Type: new Abstract: Physics-informed generative models augment a data-fitting objective with residual penalties that encourage physical constraints to hold in generated samples. However, improving constraint satisfaction need not improve fidelity to the data distribution. We study this conflict in residual-regularized diffusion models and identify two possible consequences. First, the same prediction error can incur different residual costs near different feasible states, preferring states with lower residual sensitivity (avoidance). Second, this preference can extend to feasible states outside the data support, where the residual also vanishes (leakage). Motivated by this analysis, we propose a data-anchored penalty that weights prediction errors by the residual Jacobian at the clean training sample and normalizes the resulting magnitude. This removes the incentive to relocate to less sensitive states without changing the model architecture or sampling procedure. Experiments on a controlled synthetic system, three PDE systems, and molecular transition-state generation with flow matching demonstrate the benefits of anchoring. Code is available at https://github.com/TanjaBien/anchored_diffusion.
Related
- Mitigating Data Scarcity in Spaceflight Applications for Offline Reinforcement Learning Using Physics-Informed Deep Generative Models
- Residual-augmented flow matching operators for probabilistic partial differential equations
- Gen-PINNs: Generative Adversarial Physics Informed Neural Networks for solving partial differential equations
Source: arXiv cs.LG | 2026-10-14