Papers for

semiconductor manufacturing teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Fusion improves chip defect detection by combining two anomaly methods

Efficient Patch-Based Anomaly Detection Fused with Diffusion Driven Generative Modeling for Semiconductor Wafer Bin Map Open Set Anomaly Detection

Abstract: Spatial defect signatures on wafer bin maps (WBMs) trace yield loss to specific process faults, yet supervised classifiers recognize only the defect types seen during training, and one-class detectors built on a single mechanism tend to capture either local structural deviations or global distributional violations, but rarely both. This work proposes a hybrid one-class framework that couples a patch-based student-teacher detector (EfficientAD) with a denoising diffusion probabilistic model (DDPM) used for partial-diffusion reconstruction, and fuses their percentile-calibrated scores through a fixed convex combination. Trained on only 700 normal wafers from the WM-38K mixed-type dataset and evaluated on 18,658 held-out wafers, the fused detector reached an AUROC of 0.9985 and reduced misclassifications from 852 (DDPM) and 1,412 (EfficientAD) to 618, with all pairwise differences significant at p < 0.001. Beyond aggregate accuracy, the analysis shows that the gain arises from weakly overlapping errors between the two modules, yet fixed-weight fusion recovers only 40-70% of the correction available to an oracle selector. Under the benchmark's inverted class balance, average precision and F1 saturate, while the Matthews correlation coefficient and negative predictive value expose unreliable normal predictions. Pixel-level maps further show that strong image-level separability does not imply spatial localization, and the diffusion module succeeds as a local density prior rather than through global geometric reasoning. These findings motivate sample-adaptive fusion and imbalance-aware evaluation of hybrid wafer anomaly detectors.

Wed 7 OctMachine LearningArtificial Intelligence
The gist
Detecting manufacturing defects on semiconductor wafers is hard because traditional methods either catch only small local flaws or broader unusual patterns but not both. The authors combine two approaches: one that looks at small patches of the wafer, and another that reconstructs images using a special diffusion model. By merging their outputs, the combined method spots defects more accurately than each one alone. They tested this on a large dataset of wafer images and found fewer mistakes, though they suggest smarter combination methods could improve results further.
Open → 2610.09993v1

Machine learning enhances resolution of semiconductor inspection images

Image resolution enhancement for advanced semiconductor nodes

Abstract: Advanced semiconductor nodes are pushing the limits of feature sizes and require metrology with sub-nm resolution without compromising on the throughput as needed for in-line process control. Recently, high-throughput scanning probe microscopy (SPM) based metrology and inspection tools capable of meeting these needs have been introduced to the market and qualified for use in HVM. While innovative measurement methods and tool architecture have allowed for a leap of improvement in throughput, the next step in further reducing imaging time can be obtained through the application of machine learning for enhancing the resolution of measured images for extraction of relevant parameters. In this work, we provide the general framework under which a neural network-based resolution enhancer is designed and used for SPM images. We showcase the effectiveness of this framework using measurements performed on Line/Space structures with a pitch of 200 nm. For the reusability of a pre-developed pre-trained model, we additionally leverage transfer learning and show that a new model for slightly differing structures can be re-trained and calibrated with a smaller data set of measurements performed on Line/Space structures with a pitch of 100 nm.

Mon 5 OctMachine Learning
The gist
Measuring tiny details on advanced computer chips is becoming harder and slower as the features shrink below a nanometer. The authors show how a special kind of artificial intelligence can make images from scanning probe microscopes clearer and more detailed without taking extra time to scan. They also demonstrate that a model trained on one kind of chip pattern can be quickly adapted to a slightly different pattern using fewer new measurements. This approach could speed up quality checks in chip manufacturing while keeping accuracy high.
Open → 2610.05809v1

Memory restores hidden states for better atom-by-atom crystal evolution

AtomWorld-Mem: Memory-Restored World States for Long-Horizon Atomistic Evolution

Abstract: High-fidelity atomistic evolution over long timescales requires more than observing the current crystal configuration. Instantaneous atomistic snapshots are often incomplete: locally similar configurations can correspond to different hidden dynamical contexts, future event preferences, and waiting-time scales. We argue that this snapshot ambiguity makes long-horizon atomistic evolution fundamentally a memory-based world-state restoration problem. To address this, we introduce AtomWorld-Mem, a memory-restored atomistic world model that recovers the latent world state missing from instantaneous crystal snapshots. AtomWorld-Mem treats the evolving alloy as an AtomWorld: spatial encoders write multi-scale atomistic keyframes from dense local topology and sparse long-range defect context, while short-term event memory and long-term structural memory integrate these keyframes across time to restore a future-predictive evolutionary state. The restored state is used to prioritize legal vacancy-mediated events under single-event Kinetic Monte Carlo (KMC) constraints, while event legality, physical execution, and residence-time updates remain governed by the underlying simulator. Empirically, AtomWorld-Mem improves long-horizon atomistic progress under fixed microscopic event budgets while maintaining high-fidelity evolution across energetic, structural, and vacancy-transport observables. It further transfers zero-shot across diverse unseen alloy-temperature AtomWorlds, suggesting that the learned memory-restoration mechanism captures reusable principles of hidden-state inference rather than a system-specific local energy heuristic. These results position memory-restored world-state modeling as a promising route toward efficient, physically grounded, and transferable atomistic evolution.

Fri 25 SeptArtificial Intelligence
The gist
Predicting how materials like crystals change over time is hard because looking at just one snapshot of atoms doesn’t show the full story. The authors found that keeping memory of past atom arrangements helps reveal hidden information that influences future changes. They built a system called AtomWorld-Mem that remembers previous states to improve long-term predictions of how atoms move and rearrange. This system better matches real material behavior and works well even on materials it wasn’t trained on.
Open → 2609.31133v1

Analog circuit yield improves faster with new simulation method

Simulation-Efficient Analog Circuit Yield Optimization via Monte Carlo Zeroth-Order Gradient Estimation

Abstract: Yield optimization under process variation is expensive because each candidate design must be evaluated across many Monte Carlo SPICE samples. The resulting finite-sample yield is also piecewise constant in the design parameters, providing little local information for optimization. We introduce zeroth-order Monte Carlo stochastic gradient descent (ZO-MC-SGD), a black-box method that converts continuous specification margins into stochastic descent directions. Each update evaluates opposite design perturbations under shared process samples, allowing a small simulation batch to estimate a local direction without differentiating SPICE or fitting a global surrogate model. A Spearman rank-correlation test checks that the margin-based loss orders designs consistently with empirical yield. We prove that the estimator is unbiased for a Gaussian-smoothed surrogate and derive variance and sample-complexity bounds with no explicit dependence on process dimension. Across five analog circuit benchmarks with up to 30 design variables and 42 process variables, ZO-MC-SGD reaches a mean yield of 0.95 on four circuits within 50--200 simulations and the empirical yield ceiling on the fifth. Relative to the best of five black-box and learning-based baselines, it reduces the required simulation budget by up to a factor of eight.

Fri 25 SeptComputational Engineering, Finance, and Science
The gist
Optimizing how well circuits perform despite manufacturing differences is usually very slow because each design is tested many times in computer simulations. The authors developed a new way to guess which direction improves circuit yield using fewer simulations by comparing slight design changes under the same test conditions. Their method doesn't require complicated math inside the simulation software or building approximate models. They showed it works well on several example circuits, needing fewer simulations than existing methods to reach high yield.
Open → 2609.30678v1

Morphology aware model improves wafer defect diagnosis with two options

Morphology-Aware Ambiguity Learning for Wafer Defect Decision Support

Abstract: Wafer map defect recognition is commonly formulated as a fixed-taxonomy classification problem that assigns each wafer to a single defect class. However, some wafers exhibit morphologies near class boundaries, for which forcing a single prediction may be less informative than providing plausible diagnostic alternatives. This paper proposes a morphology-aware ambiguity learning framework that supports three diagnostic actions: automatic single-class diagnosis, assisted diagnosis with two plausible defect classes, and full review. Using the radial, angular, and geometric characteristics of training wafer maps, the framework constructs a class-level ambiguity matrix representing defect-class pairs with similar morphology and plausible diagnostic alternatives. It guides the model to learn plausible alternative classes rather than treating all incorrect classes equally. During inference, the matrix determines whether an uncertain prediction can be represented by a meaningful two-class diagnostic set or should be escalated for full review. Experiments on WM-811K show that the proposed framework outperforms conventional approaches in defect recognition and diagnostic decision support, providing meaningful two-class alternatives while reserving full review for cases with unresolved ambiguity. Illustrative cost analyses further show the potential cost advantage of the proposed routing strategy. The diagnostic behavior of the framework remains consistent across different backbone architectures.

Fri 18 SeptComputer Vision and Pattern Recognition
The gist
Identifying defects on silicon wafers is usually done by assigning each wafer to one specific defect type, but sometimes wafers look like they could belong to two categories. The authors propose a method that recognizes when a wafer's defect pattern is ambiguous and offers either a single diagnosis, two plausible alternatives, or recommends a full expert review. Their approach uses the shapes and patterns of defects learned from training data to understand which classes often look similar. Tests on a large dataset show this method helps make more accurate and informative predictions, saving effort and cost by catching ambiguous cases smartly.
Open → 2609.21866v1