# US 2026/0235965 A1 · Method and system for selecting patterns based on similarity metric

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0235965 A1
- Document ID: US20260235965A1
- Status: Published application
- Kind: A1
- Application number: 19/151,445
- Filed: 2024-01-25
- Published: 2026-08-13
- Earliest claimed: 2023-02-22
- Assignees: ASML Netherlands B.V.
- Applicants: Asml Netherlands B.V.
- Inventors: Xiaogai TANG; Wei-Chou LIN
- Web page: https://patentgrep.com/patent/US20260235965A1

## CPC

- G03F 7/706837: {Data analysis, e.g. filtering, weighting, flyer removal, fingerprints or root cause analysis}
- G03F 7/706839: {Modelling, e.g. modelling scattering or solving inverse problems}
- G03F 7/706841: {Machine learning}
- G03F: PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR; (phototypographic composing devices B41B; photosensitive materials or processes for photographic purposes G03C; electrophotography, sensitive layers or processes therefor G03G)

## Abstract

Methods and systems for selecting patterns from one set to another set. A first set of patterns and a second set of patterns represented in a pattern representation space are obtained. For each pattern in the second set of patterns, a similarity metric that is indicative of a similarity between the corresponding pattern and patterns in the first set of patterns in the pattern representation space is determined. A specified pattern is selected from the second set of patterns for updating the first set of patterns based on the similarity metric.

## Claims

### Claim 1

A non-transitory computer-readable medium having instructions recorded thereon or therein, the instructions, when executed by a computer system, are configured to cause the computer system to at least:
obtain a first set of patterns and a second set of patterns represented in a pattern representation space;
obtain for each pattern in the second set of patterns, a similarity metric that is indicative of a similarity between the corresponding pattern and patterns in the first set of patterns, wherein the similarity is determined in the pattern representation space; and
select a specified pattern from the second set of patterns for updating the first set of patterns based on the similarity metric.

### Claim 2 (depends on 1)

The medium of claim 1, wherein the pattern representation space includes a feature vector space, and wherein the first set of patterns or the second set of patterns are represented in the feature vector space by encoding, based on an autoencoder model, each pattern in the first set of patterns or the second set of patterns to a corresponding feature vector.

### Claim 3 (depends on 2)

The medium of claim 2, wherein the encoding includes encoding a target layout of the first set of patterns or the second set of patterns to the corresponding feature vector.

### Claim 4 (depends on 2)

The medium of claim 2, wherein the encoding includes encoding an image of a pattern in the first set of patterns or the second set of patterns to the corresponding feature vector.

### Claim 5 (depends on 2)

The medium of claim 2, wherein the similarity metric includes distance information indicating distances in the feature vector space between the corresponding pattern in the second set of patterns and patterns in the first set of patterns.

### Claim 6 (depends on 5)

The medium of claim 5, wherein the distance information includes nearest neighbor distance between two feature vectors in the feature vector space.

### Claim 7 (depends on 5)

The medium of claim 5, wherein the instructions configured to cause the computer system to select the specified pattern are further configured to cause the computer system to do so is based on a largest nearest neighbor distance, wherein the largest nearest neighbor distance is indicative of a pattern among the second set of patterns that is least similar to any pattern of the first set of patterns.

### Claim 8 (depends on 1)

The medium of claim 1, wherein the first set of patterns includes patterns having nearest neighbor distances in the pattern representation space between other patterns in the first set of patterns above a specified threshold.

### Claim 9 (depends on 1)

The medium of claim 1, wherein the instructions configured to cause the computer system to select the specified pattern are further configured to cause the computer system to move the specified pattern from the second set of patterns to the first set of patterns.

### Claim 10 (depends on 9)

The medium of claim 9, wherein the instructions are further configured to cause the computer system to update the similarity metric for remaining patterns of the second set of patterns after movement of the specified pattern to the first set of patterns.

### Claim 11 (depends on 1)

The medium of claim 1, wherein the instructions are further configured to cause the computer system to:
iteratively select a pattern from the second set of patterns until a specified criterion is satisfied, wherein each iteration includes:
selection of a pattern of the second set of patterns based on the similarity metric;
movement of the pattern to the first set of patterns; and
update of the similarity metric for remaining patterns of the second set of patterns after movement of the pattern to the first set of patterns.

### Claim 12 (depends on 11)

The medium of claim 11, wherein the specified criterion is based on at least one selected from: (a) a number of iterations, (b) a number or percentage of patterns selected from the second set of patterns, (c) a total number of patterns in the first set of patterns, or (d) a nearest neighbor distance.

### Claim 13 (depends on 11)

The medium of claim 11, wherein the instructions configured to cause the computer system to iteratively select the pattern are further configured to cause the computer system to:
determine a pattern coverage in the first set of patterns based on a number of patterns present in the first set of patterns associated with a given similarity metric value of a set of similarity metric values with the second set of patterns; and
determine whether to continue the iterative selection based on the pattern coverage.

### Claim 14 (depends on 1)

The medium of claim 1, wherein the pattern representation space includes an image representation space,
wherein the first set of patterns or the second set of patterns are represented in the image representation space by obtaining an image of each pattern in the first set of patterns or the second set of patterns, and
wherein the similarity metric is indicative of a similarity between an image of each pattern in the second set of patterns and images of patterns in the first set of patterns.

### Claim 15 (depends on 1)

The medium of claim 1, wherein the instructions are further configured to cause the computer system to calibrate, using the first set of patterns, a model configured to determine characteristics of a patterning process or a lithography apparatus used for printing a pattern on a substrate.

### Claim 16

A non-transitory computer-readable medium having instructions recorded thereon or therein, the instructions, when executed by a computer system, are configured to cause the computer system to at least:
obtain a first set of patterns and a second set of patterns represented in a feature vector space;
obtain, for each pattern in the second set of patterns, distance information indicating distances in the feature vector space between the corresponding pattern and patterns in the first set of patterns; and
select a specified pattern from the second set of patterns for updating of the first set of patterns based on the distance information.

### Claim 17 (depends on 16)

The medium of claim 16, wherein the distance information includes nearest neighbor distance between two feature vectors in the feature vector space.

### Claim 18 (depends on 16)

The medium of claim 16, wherein the distances for the patterns in the second set of patterns are ranked from a first nearest neighbor distance to a last nearest neighbor distance.

### Claim 19 (depends on 16)

The medium of claim 16, wherein the instructions configured to cause the computer system to select the specified pattern are configured to cause the computer system to do so based on a largest nearest neighbor distance.

### Claim 20 (depends on 16)

The medium of claim 16, wherein the first set of patterns includes patterns having nearest neighbor distances between other patterns in the first set of patterns above a specified threshold.
