DETAILED ACTION
This is a request for continued Examination filed July 24, 2026.
Claims 1, 8, and 15 are amended.
Claims 5 and 12 are canceled.
Claims 2, 3, 4, 7, 9, 10, 11, 14 are originals.
Claims 6 and 13 are previously presented.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 24 July 2026 has been entered.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 1-4, 6-11, and 13-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 (The Statutory Categories): Is the claim to a process, machine, manufacture, or composition of matter? MPEP 2106.03.
Per Step 1: Claims 1-4, 6-11, and 13-15 are directed to a method, system, and non-transitory computer readable medium. Thus, each of the claims falls within one of the four statutory categories (step 1). However, the claims also fall within the judicial exception of an abstract idea (step 2). While claims 1, 8, and 15, are directed to different categories, the language and scope are substantially the same and have been addressed together below.
The analysis proceeds to Step 2A Prong One.
Step 2A Prong One: Does the claim recite an abstract idea, law of nature, or natural phenomenon? MPEP 2106.04.
The abstract idea of claim 1, 8 and 15; (Claim 1 being representative):
A method, comprising: fine-tuning a pre-trained generative language model (GLM) on a few historical records of fault codes and corresponding repair recommendations in a few-shot manner to produce a fine- tuned GLM that is fine-tuned to the equipment; for receipt of a text input requesting a recommendation for an equipment based on underlying conditions: processing the text input with the fine-tuned GLM, wherein the fine-tuned GLM leverages pre-existing knowledge encoded during initial training to enable equipment-specific optimization with fewer labeled examples of historical repairs compared to training from scratch, the fine-tuned GLM configured to output raw text representative of the recommendation for the equipment based on the underlying conditions; [[and]] processing the output raw text into a recommendation mapping process configured to map the output raw text to one or more of: a checklist of recommendations conforming to end user standard, or a list of codes related to the recommendation for the equipment conforming to the end user standard[[.]], wherein the recommendation mapping process comprises a classifier configured to output the codes in response to the output raw text based on a temporal state of the equipment, wherein the classifier is trained against historical repairs and historical temporal states of the equipment to output repair codes based on temporal state and symptom correlation, and a summarizer configured to summarize the codes into the checklist of recommendations and is configured to provide a measure of similarity to known recommendations, wherein the summarizer maps the output raw text to one or more nearest sets of valid repair recommendations based on the measure of similarity; and outputting the checklist of recommendations or the list of codes as a sequence of executable repair instructions to an automated repair system, the automated repair system executing the repair instructions on the equipment.
The abstract idea steps italicized above are those which could be performed mentally, including with pen and paper. The steps describe, at a high level, receiving, recommending, processing, storing, fine-tuning, converting, repairing and providing an output that facilitates service repair operations for the equipment. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, including observations, evaluations, judgements, and/or opinions, then it falls within the Mental Processes – Concepts Performed in the Human Mind grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
Additionally, and alternatively, the abstract idea steps italicized above relate to receiving, recommending, processing, storing, fine-tuning, converting, repairing, and providing an output that facilitates service repair operations for the equipment which constitutes a process that, under its broadest reasonable interpretation, covers commercial activity. This is further supported by [0014] of applicant’s specification as filed. If a claim limitation, under its broadest reasonable interpretation, covers commercial interactions, including contracts, legal obligations, advertising, marketing, sales activities or behaviors, and/or business relations, then it falls within the Certain Methods of Organizing Human Activity – Commercial or Legal Interactions grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two: Does the claim recite additional elements that integrate the judicial exception into a practical application? MPEP 2106.04.
This judicial exception is not integrated into a practical application because the additional elements are merely instructions to apply the abstract idea to a computer, as described in MPEP 2106.05(f).
Claim 1 recites the following additional elements: a fine-tuned generative language model (GLM), a list of codes, fault codes, fine-tuned GLM leverages pre-existing knowledge encoded during initial training to enable equipment-specific optimization with fewer labeled examples of historical repairs compared to training from scratch, the fine-tuned GLM, wherein the classifier is trained against historical repairs and historical temporal states of the equipment to output repair codes
Claim 8 recites the following additional elements: non-transitory computer readable medium, a fine-tuned generative language model (GLM), a list of codes, fault codes, fine-tuned GLM leverages pre-existing knowledge encoded during initial training to enable equipment-specific optimization with fewer labeled examples of historical repairs compared to training from scratch, the fine-tuned GLM, wherein the classifier is trained against historical repairs and historical temporal states of the equipment to output repair codes.
Claim 15 recites the following additional elements: Apparatus, processor, a fine-tuned generative language model (GLM), a list of codes, fault codes, fine-tuned GLM leverages pre-existing knowledge encoded during initial training to enable equipment-specific optimization with fewer labeled examples of historical repairs compared to training from scratch, the fine-tuned GLM, wherein the classifier is trained against historical repairs and historical temporal states of the equipment to output repair codes
These elements are merely instructions to apply the abstract idea to a computer, per MPEP 2106.05(f). Applicant has only described generic computing elements in their specification, as seen in [0014] of applicant’s specification as filed, for example.
Further, the combination of these elements is nothing more than a generic computing system applied to the tasks of the abstract idea. Because the additional elements are merely instructions to apply the abstract idea to a generic computing system, they do not integrate the abstract idea into a practical application, when viewed in combination. See MPEP 2106.05(f).
Therefore, per Step 2A Prong Two, the additional elements, alone and in combination, do not integrate the judicial exception into a practical application. The claim is directed to an abstract idea.
Step 2B (The Inventive Concept): Does the claim recite additional elements that amount to significantly more than the judicial exception? MPEP 2106.05.
Step 2B involves evaluating the additional elements to determine whether they amount to significantly more than the judicial exception itself.
The examination process involves carrying over identification of the additional element(s) in the claim from Step 2A Prong Two and carrying over conclusions from Step 2A Prong Two pertaining to MPEP 2106.05(f).
The additional elements and their analysis are therefore carried over: applicant has merely recited elements that facilitate the tasks of the abstract idea, as described in MPEP 2106.05(f).
Further, the combination of these elements is nothing more than a generic computing system. When the claim elements above are considered, alone and in combination, they do not amount to significantly more.
Therefore, per Step 2B, the additional elements, alone and in combination, are not significantly more. The claims are not patent eligible.
The analysis takes into consideration all dependent claims as well:
Dependent claim 2-4, 6-7 and 9-11 and 13-14 contain additional steps that further narrow the abstract idea above.
Claim 2 recites the following additional elements: audio input. Applicant has only described generic computing elements in their specification, as seen in {[0037]} of applicant’s specification as filed. This does not integrate the abstract idea into practical application and/or add significantly more. The claim is ineligible. Refer to MPEP 2106.05(f).
Claim 3, 4, 6, 7 recites the following additional elements: a fine-tuned generative language model (GLM), Codes. Applicant has only described generic computing elements in their specification, as seen in {[0022]} of applicant’s specification as filed. This does not integrate the abstract idea into practical application and/or add significantly more. The claim is ineligible. Refer to MPEP 2106.05(f).
Claim 9 recites the following additional elements: non-transitionary computer readable medium, a fine-tuned generative language model (GLM). Applicant has only described generic computing elements in their specification, as seen in {[0015]} of applicant’s specification as filed. This does not integrate the abstract idea into practical application and/or add significantly more. The claim is ineligible. Refer to MPEP 2106.05(f).
Claim 13 recites the following additional elements: Natural language. Applicant has only described generic computing elements in their specification, as seen in {[0022]} of applicant’s specification as filed. This does not integrate the abstract idea into practical application and/or add significantly more. The claim is ineligible. Refer to MPEP 2106.05(f).
In conclusion the claims do not provide an inventive concept, because the claims do not recite additional elements or a combination of elements that amount to significantly more than the judicial exception of the claims. Therefore, whether taken individually or as an order combination, the claims are nonetheless rejected under 35 U.S.C. 101 as being directed to non - statutory subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-4, 6-11 and 12- 15 are rejected under 35 U.S.C. 103 as being unpatentable over Luong et al [US 2022/038,3206 A1], hereafter Luong, in view of Shah et al [US 2020/041,0163 A1], hereafter Shah, in further view of Farahat et al [US 2020/025,8057 A1], hereafter Farahat.
As per claim 1, 8 and 15 (Similar scope and language);
Luong does disclose fine-tunning a pre trained generative language model, see [0028], few shot and adaptation to target [0034];
A method, comprising: fine-tuning a pre-trained generative language model (GLM) on a few historical records of fault codes and corresponding repair recommendations in a few-shot manner to produce a fine-tuned GLM that is fine-tuned to the equipment;
{[0028] Example approaches can first train an NLI data generator by fine-tuning a pre-trained generative language model on the MNLI dataset in a text-to-text format.
[0031] the techniques described herein enable models to be quickly adapted to new tasks and domains. This enables models to be fine-tuned from a starting point rather than from scratch,
[0034] Despite their recent successes in tackling many NLP tasks, large-scale pre-trained language models do not perform as well in few-shot settings where only a handful of training examples are available…. on the SST-2 sentiment dataset, STraTA, with only 8 training examples per class, achieves comparable results to standard fine-tuning with 67K training examples.}
Luong discloses the fine tuned GLM leverages pre existing knowledge encoded during initial training compared to training from the scratch; See [0031] [0037] [0045]
[0031] the techniques described herein enable models to be quickly adapted to new tasks and domains. This enables models to be fine-tuned from a starting point rather than from scratch,
[0037] example implementations can first build an NLI data generator by fine-tuning a pre-trained generative language model on the MNLI dataset in a text-to-text format.
[0045] a pre-trained language model, such as BERT, is fine-tuned on an auxiliary task before the target task.}
Luong does not explicitly disclose the following, however; Shah discloses the few historical fault codes and records, [0016] [0022] [0024] [0029]
{[0016] The historical database 118 includes historical service requests and a plurality of corresponding historical recommendations… The system logs include time stamp values, one or more message codes, fault codes and corresponding message descriptions, associated system events routinely logged by the imaging system 102 during its operation.
[0022] The third machine learning model is a multi-label classifier configured to generate one of a plurality of repair options.}
Shah discloses the text input or receiving a service request, deep-learning and classification model, recommendation based on machine condition and historical service request and recommendations; See [0004] [0014-0015] [0022] [0023] [0026];
for receipt of a text input requesting a recommendation for an equipment based on underlying conditions: processing the text input with the fine-tuned GLM, wherein the fine-tuned GLM leverages pre-existing knowledge encoded during initial training to enable equipment-specific optimization with fewer labeled examples of historical repairs compared to training from scratch, the fine-tuned GLM configured to output raw text representative of the recommendation for the equipment based on the underlying conditions;
{[0044] The method includes receiving a service request corresponding to the machine…. generating a recommendation based on the list of descriptive words and the service architecture.
[0014] The service request includes a description of a problem or a fault condition in the imaging system 102 ..…..The text processing unit 106 is configured to parse natural language text in a pre-specified language using text processing techniques.
[0015] each of the plurality of service requests includes a description of a fault or a problem faced by the user…..The description of a service request includes a list of descriptive words representative of the fault condition in the imaging system 102.
[0022] The recommendation unit 108 is configured to generate a recommendation by traversing the classification scheme based on the service architecture.
[0023] the machine learning unit 112 is configured to retrieve a plurality of historical service requests and system logs corresponding to the retrieved historical service requests.
[0026] The machine learning unit 112 is configured to generate one or more machine learning models to be used by the recommendation unit. Further, the machine learning unit 112 is configured to generate a plurality of machine learning models representative of the plurality of classification schemes based on the first plurality of binary vectors, the second plurality of binary vectors and the corresponding historical recommendations… a machine learning model, such as a neural network model, may be trained based on the first plurality of binary vectors and the second plurality of binary vectors to generate corresponding historical recommendations.}
Shah discloses processing the input into the recommendation mapping process; [0014] [0019]; mapping to one or more recommendations; see [0032]; mapping the output raw text to one or more recommendations and list of codes, see [0019]; checklist of repair recommendations, see [0034]
[[and]] processing the output raw text into a recommendation mapping process configured to map the output raw text to one or more of: a checklist of recommendations conforming to end user standard, or a list of codes related to the recommendation for the equipment conforming to the end user standard[[.]],
{[0014] The text processing unit 106 is configured to parse natural language text in a pre-specified language using text processing techniques.
[0019] the multi-label recommendation may include a service code specifying a servicing action, a part code specifying one or more parts to be replaced, an action code specifying one or more actions to be performed during the servicing.
[0032] each of the nodes specifies a plurality of processing steps to be performed on a service request to generate one or more recommendations.
[0034] a list of parts to be replaced to correct the fault of the imaging system 102…. a list of on-site repair recommendation.}
Shah discloses the recommendation process, underlying condition, classifier, classifier see [0022]; Output repair recommendations, see [0034]; Codes related to the recommendation [0019]; multiple recommendation [0034]
wherein the recommendation mapping process comprises a classifier configured to output the codes in response to the output raw text based on a temporal state of the equipment, wherein the classifier is trained against historical repairs and historical temporal states of the equipment to output repair codes based on temporal state and symptom correlation, and a summarizer configured to summarize the codes into the checklist of recommendations and is configured to provide a measure of similarity to known recommendations, wherein the summarizer maps the output raw text to one or more nearest sets of valid repair recommendations based on the measure of similarity;
{[0019] the multi-label recommendation may include a service code specifying a servicing action, a part code specifying one or more parts to be replaced, an action code specifying one or more actions to be performed during the servicing.
[0020] The text processing unit 106 is configured to parse sentences into a list of descriptive words describing the fault condition.
[0022] The recommendation unit 108 is configured to generate a recommendation by traversing the classification scheme based on the service architecture… third machine learning model is a multi-label classifier configured to generate one of a plurality of repair options.
[0034] the recommendation unit 108 may generate multiple recommendations while traversing the tree to resolve a single service request…. The fifth classification scheme is a multi-label classification scheme configured to provide a list of on-site repair recommendations.}
Shah discloses the checklist; See [0034]
{[0034] The fifth classification scheme is a multi-label classification scheme configured to provide a list of on-site repair recommendations… The recommendation unit 108 may also generate a fifth recommendation generated from the repair action 212 includes specification of a list of repair actions.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong et al’sgenerative language model to include Shah et al’s checklist, text input or receiving a service request, deep-learning and classification model, recommendation based on machine condition and historical service request and recommendations to provide a better recommendation and repair of the machine.
Luong does not explicitly disclose the following, however; Farahat discloses the ML system organizes /determines repair actions and corresponding repair plans that include the steps for implementing the respective repair actions; See [0075]
{[0075] the computing device may learn or otherwise determine repair plans 221 corresponding to the possible repair actions that may be output by the trained machine learning model(s)… the steps taken for implementing each different repair action that may be determined by the machine learning model.}
Farahat discloses, the remote automated execution; See [0040 - 0041]
and outputting the checklist of recommendations or the list of codes as a sequence of executable repair instructions to an automated repair system, the automated repair system executing the repair instructions on the equipment.
{[0040] the repair plan execution program 136 may provide repair instructions 162 to a client computing device 110, such as the repairer computing device
[0041] the repair instructions 162 may cause the equipment itself to perform a repair or otherwise initiate a repair… the repair application 118 on the equipment computing device 116 may receive the repair instructions 162 and may perform one or more operations in accordance with the repair instructions 162 for performing the repair….the repair instructions 162 may include computer executable instructions that may be executed by the equipment computing device 116 for performing or otherwise initiating the repair….the repair plan execution program 136 may initiate remotely a repair operation via the equipment computing device 116 for performing a repair on the corresponding equipment 152.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combine/modify/adjust Luong et al’sgenerative language model to include Farahat et al’s ML system that organizes /determines repair actions and corresponding repair plans and remote automated execution to provide repair instructions and summaries.
As per Claim 2 and 9 (Similar scope and language);
The combination of Luong, Farahat does not disclose the following, however; Shah discloses;
The method of claim 1, wherein the text input is converted from audio input from the end user.
{[0013] The service request may be a description in text or audio message provided by the user via a user interface and may be automatically stored in a database}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combine/modify/adjust Luong et al’sgenerative language model to include Shah et al’s text or audio message to process a speech-to-text.
As per Claim 3 and 10 (Similar scope and language):
The combination of Luong, Farahat does not disclose the following, however; Shah discloses;
The method of claim 1, further comprising processing the text input with a pre-processing layer configured to remove edge cases, vagueness, and implicitness from the text input and is further configured to formulate the text input into a specific problem and repair case of the equipment for input into the fine-tuned GLM.
{[0020] The text processing unit 106 is configured to parse sentences into a list of descriptive words describing the fault condition. The text processing unit 106 is configured to remove special characters, proper nouns and dates from the service request description.
[0027] The text processing includes removing special characters, proper nouns and dates from the service request to generate the list of descriptive words.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong et al’sgenerative language model to include Shah removing special characters, proper nouns and dates to enable the formulation of the text input.
Luong discloses the pre-trained generative language model;
{[0028] Example approaches can first train an NLI data generator by fine-tuning a pre-trained generative language model on the MNLI dataset in a text-to-text format.}
As per Claim 4 and 11 (Similar scope and language);
The combination of Luong and Shah does not disclose the following, however; Farahat discloses the data used in training of the Machine learning Model
The method of claim 1, wherein the fine-tuned GLM is curated with structured and unstructured data related to the equipment.
{[0017] The system may train at least one machine learning model using the at least one feature and the plurality of other features.}
Farahat discloses the structured and unstructured data
{[0017] The system may identify and extract free-form text in the historical repair data, such as words, phrases, topics, or other free-form text.
[0046] The training data obtained from the historical repair data and/or the data included in a repair request received from a client computing device may include or may be associated with a plurality of different types of data such as equipment attributes, equipment usage data, equipment sensor data, event data, user complaints and/or error messages, other complaint-related data, a repair history for the equipment and/or the equipment model, and various types of metadata.
[0047] The equipment attributes may be structured data that encodes the attributes of the equipment to be repaired.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong and Shah et al’s generative language model to include Farahat structured and unstructured data to enable obtaining the historical repair data.
As per Claim 6 and 13 (Similar scope and language);
The combination of Luong and Shah does not disclose the following, however; Farahat discloses structured non-textual equipment data [0047]; textual fault [0051], combining structured and unstructured equipment, see [0091];
The method of claim 1, further comprising converting structured, non-textual equipment data into natural language descriptions and combining with textual fault codes for input to the fine-tuned GLM, wherein the GLM produces raw text repair recommendations from both structured and unstructured data.
{[0047] The equipment attributes may be structured data that encodes the attributes of the equipment to be repaired… the make and model of the equipment, the manufacturing year of the equipment, and the capacity, ratings, or other quantifiable characteristics of the equipment and/or the equipment components.
[0051] The user complaints and error messages may be natural language complaints received from the equipment user as well as any error messages, error codes, fault codes, or the like, received from the equipment or other systems in the environment of the equipment…. unstructured or semi-structured data that describe the symptoms of the failure
[0091] the complaints and error messages 316 may include free text as well as error codes or the like.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong and Shah et al’s generative language model to include Farahat combination of structured and unstructured data to describe the symptoms of the failure in simple terms.
The combination of Luong and Farahat does not disclose the following, however; Shah discloses the repair recommendations, See [0023] [0024] [0026].
{[0023] The machine learning unit 112 is further configured to generate the service dictionary, the part dictionary and the plurality of classification schemes that are used to process the service request to generate the recommendation.
[0024] the machine learning unit 112 is configured to control the text processing unit 106 to initiate parsing of the retrieved historical service requests to generate a list of historical descriptive key-words. Further, the machine learning unit 112 is configured to control the text processing unit 106 to initiate parsing of historical system logs corresponding to the historical service requests to generate a list of historical message codes. The machine learning unit 112 is further configured to establish an association between the plurality of historical descriptive key-words with the plurality of historical message codes to generate the service dictionary.
[0026] The machine learning unit 112 is configured to generate one or more machine learning models to be used by the recommendation unit.}
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong et al’sgenerative language model to include Shah repair recommendations to enable repair recommendations.
Luong discloses the input to the fine-tuned GLM, See [0096];
{[0096] the input to the machine-learned model(s) of the present disclosure can be text or natural language data.}
As per Claim 7 and 14 (Similar scope and language);
The combination of Luong and Shah does not disclose the following, however; Farahat discloses the determination of repair plans/actions, steps to implement the repair, sequence used to conduct the recommendation, [0075-0076]; equipment performing repair, See [0041].
The method of claim 1, wherein the checklist of recommendations or the list of codes is output according to a sequence to conduct the recommendations or the list of codes.
[0075] the computing device may learn or otherwise determine repair plans 221 corresponding to the possible repair actions that may be output by the trained machine learning model(s) 219… the repair plan execution program may learn from the historical repair data, or other sources, the steps taken for implementing each different repair action that may be determined by the machine learning model.. the repair plan execution program may obtain, e.g., from the extracted data or other data sources, the repair steps that are indicted to be used for performing the repair.
[0076] Based on the learned repair steps for each repair action, subsequently, during the model application stage, the repair plan execution program may, in some examples, execute one or more of the aforementioned steps using the learned repair plan.
[0041] the repair instructions 162 may cause the equipment itself to perform a repair or otherwise initiate a repair. For example, the repair application 118 on the equipment computing device 116 may receive the repair instructions 162 and may perform one or more operations in accordance with the repair instructions 162 for performing the repair…. the repair instructions 162 may include computer executable instructions that may be executed by the equipment computing device 116 for performing or otherwise initiating the repair.
Motivation: It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combined/modify/adjust Luong and Shah et al’s generative language model to include Farahat determination of repair plans/actions, steps to implement the repair, sequence used to conduct the recommendation to enable repair instructions.
Response to Arguments
In response to the argument filled July 24, 2026, on page 8-10, regarding the 101 rejections, the Examiner Respectfully disagrees.
Applicant states that the claims are technical improvements “adapt a generative language model (GLM) to recommend repairs for an equipment based on a set of user complaints received by the user along with a set of fault codes from that equipment”. Applicant further argues that the claimed invention provides an improvement to how machine learning model operates through equipment specific fine-tuning and the recommendation mapping process ensures valid repair recommendations by "mapping the raw output to one or more nearest sets of repair recommendations based on a measure of similarity."
The Examiner respectfully disagrees.
The Examiner notes that the unstructured raw text GLM outputs, the structured mapping process, valid recommendations, the automated repair process for the equipment, the machine learning model, are merely generic technology with no technical improvement, rather an improvement to a practical application that further provides a solution for a computer-specific problem; which is a form of an abstract idea using generic technology. See specification [0029] [0054].
Applicant states that the claims are as amended, reflect improvements by reciting: (1) an affirmative fine- tuning step that fine-tunes a pre-trained GLM on a few historical records of fault codes and corresponding repair recommendations in a few-shot manner; (2) a fine-tuned GLM that leverages pre-existing knowledge to enable equipment-specific optimization with fewer labeled examples; (3) a classifier trained against historical repairs and historical temporal states to output repair codes based on temporal state and symptom correlation; (4) a summarizer that maps output raw text to nearest sets of valid repair recommendations based on similarity measures; and (5) outputting the recommendations as a sequence of executable repair instructions to an automated repair system, the automated repair system executing the repair instructions on equipment.
The Examiner respectfully disagrees.
The Examiner notes that fine- tuning a GLM to recommend, optimize, classify, output, summarize the map output raw text, instruct and execute repair instructions are merely generic technology with no technical improvement rather an improvement to a practical application that further provides a solution for a computer-specific problem using a trained GLM; which is a form of an abstract idea using generic technology.
The Examiner further notes that pre-training a neural network, perform a backpropagation to fine-tune a GLM on equipment-specific data, train a classifier against historical repairs and historical temporal states of equipment to output repair codes based on temporal state and symptom correlation, compute similarity measures to map raw text to nearest valid repair recommendations are all directed to a mental process.
The Examiner notes that the aspect pertaining to the fine-tuning, converting, mapping, codes, and provision of an output of raw text, the examiner viewed as steps of the identified abstract idea in the Step 2A Prong 1 Analysis and the codes as an additional element in the Step 2A Prong 2 Analysis. Therefore, the examiner maintains the claims recite Certain Methods of Organizing Human Activity – Commercial or Legal Interactions grouping of abstract ideas. See Applicant specification [0014].
The Examiner maintains these claims recite an abstract idea.
Therefore, for the foregoing reasons the Examiner has maintained the 35 USC 101 rejection.
In response to the amended claims, Examiner includes the prior art of Luong et al, in view of Shah et al, in further view of Farahat et al.
Applicant’s arguments with respect to claim(s) 1-4, 6-11, and 12- 15 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Ghosh et al (US 2020/009,7921 A1)- directed towards a computer system receiving historical repair data for first equipment, and extracting features from the historical repair data for the first equipment as training data including one or more of: free-text variables associated with comments related to the first equipment; usage attributes associated with the first equipment; equipment attributes associated with the first equipment; sensor data associated with the first equipment; or event data associated with the first equipment. The system may determine a repair hierarchy including a plurality of repair levels for the equipment. The system may use the training data to train a machine learning model as a multilayer model trained to perform multiple tasks for predicting individual levels of the repair hierarchy. The system may receive a repair request associated with second equipment and uses the machine learning model to determine at least one repair action based on the received repair request.
Russo AG, Ciarlo A, Ponticorvo S, Di Salle F, Tedeschi G, Esposito F. Explaining neural activity in human listeners with deep learning via natural language processing of narrative text. Sci Rep. 2022 Oct 25;12(1):17838.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VICTOR CHIGOZIRIM ESONU whose telephone number is (571)272-4883. The examiner can normally be reached Monday - Friday 9:00 am - 5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, SARAH MONFELDT can be reached on (571) 270-1833. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, vis it: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/VICTOR CHIGOZIRIM ESONU/
Examiner, Art Unit 3629
/SARAH M MONFELDT/Supervisory Patent Examiner, Art Unit 3629
