Physics-informed generative AI for semiconductor manufacturing: Enforcing hard physical constraints in generative models by construction

arXiv cs.LG Papers
physics-informed generative-ai semiconductor-manufacturing diffusion-models physical-constraints differentiable-lithography research-agenda

Summary

This paper argues that generative AI for semiconductor manufacturing must enforce hard physical constraints by construction, not via post-hoc filtering, and surveys architectural approaches like physics-informed diffusion and neural-operator priors to achieve physics fidelity.

arXiv:2606.11247v1 Announce Type: new Abstract: Generative models are increasingly used to propose designs, data, and control actions for physical systems, yet many such systems are governed by hard physical constraints rather than by perceptual plausibility. Semiconductor manufacturing provides a demanding test case: generated masks, layouts, synthetic defect data, and process recipes must obey lithography, transport, reaction, and device-physics constraints, because physically invalid samples are not merely low quality but unusable. This Perspective argues that semiconductor manufacturing exposes a broader computational-science challenge, namely that generative AI for constrained physical domains must be physics-informed by construction, not corrected only through post-hoc filtering. We survey the emerging architectural toolkit, including physics-informed diffusion, PDE-constrained variational models, neural-operator priors, and conservation-law-respecting generative networks, and show how it connects to differentiable lithography, TCAD, process simulation, and autonomous experimentation. We identify four integration patterns between generative models and physics-based simulators, and we propose a research agenda centered on physics-fidelity benchmarks, differentiable simulator infrastructure, and multimodal foundation models for physical design and manufacturing. The central claim is analytical rather than rhetorical: where physical validity is the binding criterion of success, architectures that enforce it by construction should be expected to outperform those that filter for it after the fact, and the fab is the setting where this distinction is sharpest.
Original Article
View Cached Full Text

Cached at: 06/11/26, 01:45 PM 

# Physics-informed generative AI for semiconductor manufacturing: Enforcing hard physical constraints in generative models by construction
Source: [https://arxiv.org/abs/2606.11247](https://arxiv.org/abs/2606.11247)
[View PDF](https://arxiv.org/pdf/2606.11247)

> Abstract:Generative models are increasingly used to propose designs, data, and control actions for physical systems, yet many such systems are governed by hard physical constraints rather than by perceptual plausibility\. Semiconductor manufacturing provides a demanding test case: generated masks, layouts, synthetic defect data, and process recipes must obey lithography, transport, reaction, and device\-physics constraints, because physically invalid samples are not merely low quality but unusable\. This Perspective argues that semiconductor manufacturing exposes a broader computational\-science challenge, namely that generative AI for constrained physical domains must be physics\-informed by construction, not corrected only through post\-hoc filtering\. We survey the emerging architectural toolkit, including physics\-informed diffusion, PDE\-constrained variational models, neural\-operator priors, and conservation\-law\-respecting generative networks, and show how it connects to differentiable lithography, TCAD, process simulation, and autonomous experimentation\. We identify four integration patterns between generative models and physics\-based simulators, and we propose a research agenda centered on physics\-fidelity benchmarks, differentiable simulator infrastructure, and multimodal foundation models for physical design and manufacturing\. The central claim is analytical rather than rhetorical: where physical validity is the binding criterion of success, architectures that enforce it by construction should be expected to outperform those that filter for it after the fact, and the fab is the setting where this distinction is sharpest\.

## Submission history

From: Yaser Banad \[[view email](https://arxiv.org/show-email/fd9ffae5/2606.11247)\] **\[v1\]**Mon, 8 Jun 2026 07:06:11 UTC \(1,098 KB\)

Similar Articles

Physics-IQ Verified

Hugging Face Daily Papers

This paper presents a systematic audit of the Physics-IQ benchmark for evaluating physical understanding in video generative models, proposing improvements to prompts and scoring to enhance reliability.

Synthics: Synthetic Physics-like Datasets for Machine Learning

arXiv cs.LG

A method using Bayesian Probabilistic Context-Free Grammar to generate synthetic regression datasets that structurally resemble physics equations, validated against the Feynman corpus and shown to be effective for hyperparameter tuning.

Building The Ph(ysical)AI Layer Of Machine Intelligence

arXiv cs.LG

Researchers at MIT Lincoln Laboratory propose 'principle-driven foundation models' that encode signal-theoretic physical principles (Fourier decomposition, energy conservation, symmetry) instead of learning statistical correlations from large paired datasets. Trained exclusively on RF data, their 1.99M parameter frozen encoder achieves 77.7% average accuracy across 15 diverse tasks spanning audio, images, text, and video without any fine-tuning on target domains.