DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This final office action is in response to the amendment filed 31 August 2026.
Claims 1-20 are pending. Claims 1, 11, and 18 are independent claims.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 4-6, 8-9, 11, 13-14, 16-18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Poulis et al. (US 2025/0285015, published 11 September 2025, hereafter Poulis) and further in view of Minkin et al. (US 2018/0165604, published 14 June 2018, hereafter Minkin) and further in view of Kashyap et al. (US 2025/0217214, filed 28 December 2023, hereafter Kashyap).
As per independent claim 1, Poulis discloses a computing system comprising:
one or more processors (Figure 1; paragraph 0031)
one or more non-transitory, computer-readable storage media storing instructions that, when executed by the one or more processors (Figure 1; paragraph 0033), cause the computing system to:
obtain a first input comprising (1) an input prompt associated with a domain, (2) a domain-specific key-value dataset, and (3) a domain-specific ontology map (Figure 4, item 404; paragraph 0053)
wherein the domain specifies a subset of knowledge vector space (Figure 4, item 402; paragraphs 0037 and 0052: Here, each sentence is vectorized as it is ingested by the model to determine scores representing each sentence and article as a whole)
wherein the domain-specific key-value dataset includes a key set representing a set of textual flags, associated with the domain, and a value set including a set of natural language descriptions (Figure 4, item 402; paragraphs 0037 and 0052: Here, each set of domain specific data may be divided into a key set. The key being a specific sentence in the content. This key is scored/rated based on elements such as: clickbait, exaggeration, subjectivity, source quality, dog whistle detection, and/or political bias. These scores (values) are associated at the sentence (key) level and at the article level)
wherein each key of the key set corresponds to a particular value of the value set (Figure 4, item 402; paragraphs 0037 and 0052: Here, each set of domain specific data may be divided into a key set. The key being a specific sentence in the content. This key is scored/rated based on elements such as: clickbait, exaggeration, subjectivity, source quality, dog whistle detection, and/or political bias. These scores (values) are associated at the sentence (key) level and at the article level)
wherein the domain-specific ontology map comprises a node set associated with the domain and a node relationship set between one or more nodes of the node set (paragraph 0053: Here, a domain alignment knowledge graph is generated (ontology map). This knowledge graph is used to verify that the LLM/LMM is in alignment with the principles of the specific domain determined by the query posed by the user)
determine a version of the input prompt (Figure 4, item 416; paragraph 0055: Here, a user submits a prompt/query and the embeddings are used to search for results that are sent to the LLM/LMM to generate candidate responses)
input the version of the input prompt (Figure 4, item 416), the domain-specific key-value dataset (Figure 4, item 418), and the domain-specific ontology map (paragraph 0053) into a domain-sensitive language model (Figure 4, item 412) to generate a first output set comprising a first natural language data set responsive to the version of the input prompt and associated with the domain (Figure 4, item 413: paragraphs 0052-0056: Here, domain specific data is provided for ingestion and extension of the pre-trained LLM/LMM. Based upon the prompt/query provided by a user, the prompt is provided to the pre-trained LLM/LMM (Figure 4, item 412) and the embedding models (Figure 4, item 418). Results from the vector database (Figure 4, item 106B), the embedding models (Figure 4, item 418), and retrieved results (Figure 4, item 422) are used to generate the one or more candidate responses (Figure 4, item 413) corresponding to the domain)
input the domain-specific ontology map and the first output set into an ontology validation model to generate a first validation output set associated with the first output set (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles of the domain and ranked)
wherein the first validation output set indicates whether the first output set is consistent with a rule set derived from the domain-specific ontology map (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles of the domain and ranked)
wherein the rule set includes at least one of: (1) structural rules, (2) semantic rules, (3) domain rules, or (4) compliance rules derived from the domain-specific ontology map (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles (rules) of the domain and ranked)
in response to determining that the first validation output set indicates that the first output set is consistent with the rule set, determine a validated output set comprising the first output set (paragraph 0056: Here, based upon the determination that the candidate responses align with the principles of the domain, the responses are scored and ranked. These candidate responses may then be used to retrain (tune) the model. Further, the final response may be provided responsive to the user query)
in response to generating the validated output set, input the input prompt into a base language model to determine an unconstrained output set comprising natural language data responsive to the input prompt (Figure 4; paragraphs 0055-0056: Here, the candidate responses that have been confirmed as complying with the principles of the domain constitute a validated output set)
provide the validated output set and the unconstrainted output set to a domain evaluation model to generate a domain-specificity indicator indicating whether the validated output set is specific to the domain based on a comparison between the validated output set and the unconstrained output set (paragraph 0056: Here, based upon the determination that the candidate responses align with the principles of the domain, the responses are scored and ranked. In this instance, the scoring of the output sets based upon a domain evaluation model provide an indicator indicating whether the unconstrained output set is specific to a domain)
in response to determining that the validated output set is specific to the domain, transmit the validated output set to a user device associated with the domain (paragraph 0056: Here, based upon the determination that the candidate responses align with the principles of the domain, the responses are scored and ranked. These candidate responses may then be used to retrain (tune) the model. Further, the final response may be provided responsive to the user query)
Poulis fails to specifically disclose:
wherein the ontology map comprises, in a computational logic ontology language format
wherein the validation output set comprises structured data indicating a logical inconsistency, a suggest prompt correction, and at least one of (1) violated rules of the rule set that are violated by the first output set or (2) a location of a violation within the first output set
in response to determining that the first validation output set indicates that the first output set is inconsistent with the rule set, generate one or more modified versions of the input prompt based on the first validation output set and including the suggested prompt correction
iteratively input the one or more modified versions of the input prompt into the domain-sensitive language model to generate an updated output set that is consistent with the rule set of the domain-specific ontology map
wherein the validated output set, generated by the domain-sensitive language model, is consistent with the rule set derived from the domain-specific ontology map
However, Minkin, which is analogous to the claimed invention because it is directed toward machine learning, discloses wherein the ontology map comprises, a computational logic ontology language format (paragraph 0102: Here, an ontology may be represented in Web Ontology Language (OWL)). It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Minkin with Poulis, with a reasonable expectation of success, as it would have allowed for importing ontology data using a standard for semantic representation (Minkin: paragraph 0102).
Additionally, Kashyap, which is analogous to the claimed invention because it is directed toward rule creation, discloses:
wherein the validation output set comprises structured data indicating a logical inconsistency, a suggest prompt correction, and at least one of (1) violated rules of the rule set that are violated by the first output set or (2) a location of a violation within the first output set (paragraph 0130-0133: Here, a selected component or rule is validated to determine whether it “makes sense” in the context of platform. The examiner interprets this as identifying a logical inconsistency. Based upon this inconsistency, a user is prompted to revise the input. This prompt includes an indication that a trigger is problematic. The examiner interprets this as identifying a location of the violation)
in response to determining that the first validation output set indicates that the first output set is inconsistent with the rule set, generate one or more modified versions of the input prompt based on the first validation output set and including the suggested prompt correction (paragraph 0133: Here, a proposed solution (modified version) is generated and displayed to a user to confirm/reject)
iteratively input the one or more modified versions of the input prompt into the domain-sensitive language model to generate an updated output set that is consistent with the rule set (paragraph 0076: Here, the prompt templates are iteratively submitted to the generative output service)
wherein the validated output set, generated by the domain-sensitive language model, is consistent with the rule set derived (paragraphs 0130-0133: Here, it is determined that the component or rules is valid)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Kashyap with Poulis-Minkin, with a reasonable expectation of success, as it would have allowed for validating, and in some instances modifying a prompt, in order to improve the prompt (Kashyap: paragraph 0133).
As per dependent claim 4, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Poulis discloses wherein the version of the input prompt comprises the input prompt (Figure 4, item 416; paragraph 0055: Here, a user submits a prompt/query and the embeddings are used to search for results that are sent to the LLM/LMM to generate candidate responses).
As per dependent claim 5, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Poulis discloses wherein the instructions for generating the domain-specific indicator cause the computing system to:
determine, using the domain-specific ontology map, a domain-specific characteristic set characterizing natural language datasets associated with the domain (Figure 4; paragraphs 0055-0056: Here, candidate responses (natural language datasets) are generated based upon the domain specific data and user prompt)
wherein the domain-specific characteristic set includes at least one of a domain-specific structural characteristic, a domain-specific semantic characteristic, or a domain-specific lexical characteristic (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles (rules) of the domain and ranked)
determine a base characteristic metric value set for the unconstrained output set (paragraphs 0055-0056: Here, a score is calculated (metric value) and candidate responses are ranked based upon compliance with domain principles)
wherein each base characteristic metric value of the base characteristic metric value set indicates a particular degree of compliance of the unconstrained output set with a particular domain-specific characteristic of the domain-specific characteristic set (paragraphs 0055-0056: Here, a score is calculated (metric value) and candidate responses are ranked based upon compliance with domain principles)
determine a characteristic metric value set for the validated output set (paragraphs 0055-0056: Here, a score is calculated (metric value) and candidate responses are ranked based upon compliance with domain principles)
wherein each characteristic metric value of the characteristic metric value set indicates a particular degree of compliance of the validated output set with a particular domain-specific characteristic of the domain-specific characteristic set (paragraphs 0055-0056: Here, a score is calculated (metric value) and candidate responses are ranked based upon compliance with domain principles)
compare the base characteristic metric value set with the characteristic metric value set (paragraphs 0055-0056: Here, a score is calculated (metric value) and candidate responses are ranked based upon compliance with domain principles. This ranking inherently includes a comparison to determine the highest ranking candidate response)
in response to comparing the base characteristic metric value set with the characteristic metric value set, generate the domain-specificity indicator including an indication that the validated output set is specific to the domain (paragraph 0056: Here, based upon the determination that the candidate responses align with the principles of the domain, the responses are scored and ranked. In this instance, the scoring of the output sets based upon a domain evaluation model provide an indicator indicating whether the unconstrained output set is specific to a domain)
As per dependent claim 6, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 5, and the same rejection is incorporated herein. Poulis discloses wherein the instructions for comparing the base characteristic metric value set with the characteristic metric value set causes the computing system to determine a different metric value characterizing a difference between the characteristic metric value set and the base characteristic metric value set (paragraph 0056: Here, based upon a determination that the labeling functions are “weak,” the LLM/LMM is retrained/tuned to improve responses with respect to the domain principles).
Poulis fails to specifically disclose determining that the difference metric value exceeds a threshold difference metric value. However, the examiner takes official notice that it was notoriously well-known in the art at the time to compare a value to a threshold. It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Poulis’s teaching of calculate scores and identifying labels as “weak” labels with the well-known comparison of scores to thresholds, as it would have allowed for implementing a threshold metric for identifying the strength of labels. This would have provided the advantage of maintaining consistency in labeling items and triggering retraining/tuning.
As per dependent claim 8, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated here. Poulis discloses wherein the instructions for obtaining the domain-specific ontology map further cause the computing system to:
obtain a preliminary domain-specific ontology map (Figure 4; paragraphs 0052-0056: Here, a domain-specific ontology map is provided via a knowledge graph. This is used to generate a plurality of candidate responses in conjunction with the LLM/LMM)
obtain a representation of a domain-related update (paragraph 0056: Here, based upon the candidate responses and/or feedback, the LLM/LMM may be retrained/tuned to improve responses)
wherein the domain-related update includes a modification in at least one of an ontological constraint, validation criterion, a conceptual hierarchy attribute, or a reasoning rule set (paragraph 0056: Here, based upon the candidate responses and/or feedback, the LLM/LMM may be retrained/tuned to improve responses)
provide the representation of the domain-related update and the preliminary domain-specific ontology map to the base language model to generate an updated domain-specific ontology map comprising the domain-specific ontology map (paragraph 0056: Here, based upon the candidate responses and/or feedback, the LLM/LMM may be retrained/tuned to improve responses)
As per dependent claim 9, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Poulis discloses wherein the instructions for obtaining the domain-specific ontology map further cause the computing system to:
obtain, via a graphical user interface of the user device (paragraph 0028), a representation of a domain-specific ontology comprising at least one of: a class hierarchy dataset, an object property dataset, a cardinality constraint dataset, a range constraint dataset, a complex logical expression dataset, or a computational logic ontology language rule set (paragraphs 0053-0054: Here, the domain alignment data ingestion (Figure 4, item 406) receives domain specific data (Figure 4, item 402) to generate a knowledge graph to align the LLM/SMM with the principles of the specific domain)
provide the representation of the domain-specific ontology to a computational reasoner model to generate an inferred ontological rule set associated with the representation of the domain-specific ontology (paragraphs 0053-0054: Here, the domain alignment data ingestion (Figure 4, item 406) receives domain specific data (Figure 4, item 402) to generate a knowledge graph to align the LLM/SMM with the principles of the specific domain. In this instance, the knowledge graph is the ontological map containing inferred ontological rule set associated with the representation of the domain-specific ontology)
generate the domain-specific ontology map based on the inferred ontological rule set (paragraphs 0053-0054: Here, the domain alignment data ingestion (Figure 4, item 406) receives domain specific data (Figure 4, item 402) to generate a knowledge graph to align the LLM/SMM with the principles of the specific domain)
With respect to independent claims 11 and 18, the claim recites the limitations substantially similar to those in claim 1. Claims 11 and 18 are rejected under similar rationale.
Further, Poulis discloses an artificial intelligence model (paragraph 0002: Here, the artificial intelligence model is a large language model/large multimodal model).
With respect to claims 13 and 20, the claims recite the limitations substantially similar to those in claim 5. Claims 13 and 20 are rejected under similar rationale.
With respect to claim 14, the claims recite the limitations substantially similar to those in claim 6. Claim 14 is rejected under similar rationale.
With respect to claim 16, the claims recite the limitations substantially similar to those in claim 8. Claim 16 is rejected under similar rationale.
With respect to claim 17, the claims recite the limitations substantially similar to those in claim 9. Claim 17 is rejected under similar rationale.
Claims 2-3, 12, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Poulis, Minkin, and Kashyap and further in view of Padgett et al. (US 2024/0160902, published 16 May 2024, hereafter Padgett).
As per dependent claim 2, Poulis, Minkin, and Kashyap disclose the limitations substantially similar to those in claim 1, and the same rejection is incorporated herein. Poulis further discloses wherein the instructions for determining the version of the input prompt cause the computing system to:
input the input prompt, the domain-specific key-value dataset and the domain-specific ontology map into the domain-sensitive language model to generate a preliminary output set comprising preliminary natural language data responsive to the input prompt (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles of the domain and ranked)
input the domain-specific ontology map and the preliminary output set into the ontology validation model to generate a preliminary validation output set associated with the preliminary output set (paragraph 0056: Here, based upon the determination that the candidate responses align with the principles of the domain, the responses are scored and ranked. In this instance, the scoring of the output sets based upon a domain evaluation model provide an indicator indicating whether the unconstrained output set is specific to a domain)
wherein the preliminary validation output set indicates that the preliminary output set is inconsistent with the rule set derived from the domain specific ontology map (paragraph 0056: Here, based upon a determination that the labeling functions are “weak,” the LLM/LMM is retrained/tuned to improve responses with respect to the domain principles. This label indicates that the output set is inconsistent with the principles (rule set) of the domain)
Poulis fails to specifically disclose generating, using at least one of the base language model or the domain-sensitive language model, the version of the input prompt comprising a modified version of the input prompt.
However, Padgett, which is analogous to the claimed invention because it is directed toward modifying prompts, discloses generating, using at least one of the base language model or the domain-sensitive language model, the version of the input prompt comprising a modified version of the input prompt (Figure 2; paragraphs 0076-0077 and 0083: Here, a generative AI generates a result set based upon a prompt. The result set is analyzed to determine if it is too similar/too dissimilar based upon a threshold value (paragraph 0083). In this instance, the prompt may be modified and a new result set generated (paragraph 0084)).
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Padgett with Poulis-Minkin, with a reasonable expectation of success, as it would have allowed for modifying a prompt to perform multiple iterations of generation until the results meet the predefined threshold (Padgett: paragraph 0089).
As per dependent claim 3, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Poulis discloses wherein the instructions for determining the version of the input prompt cause the computing system to:
obtain a latest version of the set of previous input versions of the input prompt (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles of the domain and ranked. In this instance, the provided version is the latest version in a set consisting of at least one version)
wherein each version of the set of previous versions is input into the domain-sensitive language model to generate a corresponding preliminary output set that is used to generate a subsequent version of the set of previous versions (Figure 4, item 410; paragraphs 0055-0056: Here, the LLM/LMM generates a plurality of candidate responses. These responses are validated using a knowledge graph to insure alignment with the principles of the domain and ranked. In this instance, the provided version is the latest version in a set consisting of at least one version)
Poulis fails to specifically disclose:
determine an iteration count representing a number of iterations associated with the set of previous input version
compare the iteration count with a threshold iteration count representing a predetermined number of allowed iterations
in response to determining that the iteration count is equal to the threshold iteration count, determine that the version of the input prompt corresponds to the latest version of the set of previous input versions
However, Padgett, which is analogous to the claimed invention because it is directed toward generating prompts, discloses:
determine an iteration count representing a number of iterations associated with the set of previous input version (Figure 3A; paragraphs 0091-0095: Here, an iteration (index) count i is used to represent the number of times a prompt has been modified and results generated)
compare the iteration count with a threshold iteration count representing a predetermined number of allowed iterations (Figure 3A; paragraphs 0091-0095: Here, it is determined that iteration (index) count is zero, the result having the lowest similarity measure is selected)
in response to determining that the iteration count is equal to the threshold iteration count, determine that the version of the input prompt corresponds to the latest version of the set of previous input versions (Figure 3A; paragraphs 0091-0095: Here, it is determined that iteration (index) count is zero, the result having the lowest similarity measure is selected)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Padgett with Poulis-Minkin, with a reasonable expectation of success, as it would have allowed for selecting the best result from a plurality of iterations of a prompt (Padgett: paragraph 0095).
With respect to claims 12 and 19, the claims recite the limitations substantially similar to those in claim 3. Claims 12 and 19 are rejected under similar rationale.
Claims 7 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Poulis, Minkin, and Kashyap and further in view of Schwartz et al. (US 2026/0147810, filed 27 November 2024, hereafter Schwartz).
As per dependent claim 7, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 1, and the same rejection is incorporated herein. Poulis discloses input at least one of the domain-specific ontology map or the domain-specific key-value dataset into the base language model (paragraph 0053: Here, a domain alignment knowledge graph is generated (ontology map). This knowledge graph is used to verify that the LLM/LMM is in alignment with the principles of the specific domain determined by the query posed by the user), the prompt includes at last one of: (1) domain-specific structural information, (2) domain-specific validation criteria, (3) domain-specific natural language constraint rules, or (4) a domain-specific example natural language dataset (paragraphs 0055-0056).
Poulis fails to specifically disclose:
generate a generation template for output generation based on the input prompt
generate an augmented input prompt including a representation of the generation template
input the augmented input prompt into the base language model to generate the first output set
However, Schwartz, which is analogous to the claimed invention because it is directed toward utilizing a prompt template with a large language model, discloses: generate a generation template for output generation based on the input prompt (paragraphs 0055-0056: Here, a prompt template is generated to maintain a prompt chain)
generate an augmented input prompt including a representation of the generation template (paragraphs 0055-0056: Here, a prompt template is generated to maintain a prompt chain by customizing (augmenting) input prompts)
input the augmented input prompt into the base language model to generate the first output set (paragraphs 0055-0056: Here, a prompt template is generated to maintain a prompt chain. The various customized templates are inputted to generate output sets)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Schwartz with Poulis-Minkin, with a reasonable expectation of success, as it would have allowed for generating prompt changes for retaining and utilizing context (Schwartz: paragraph 0055).
With respect to claim 15, the claims recite the limitations substantially similar to those in claim 7. Claim 15 is rejected under similar rationale.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Poulis, Minkin, and Kashyap and further in view of Panwar (US 2025/0077782, published 6 March 2025).
As per dependent claim 10, Poulis, Minkin, and Kashyap disclose the limitations similar to those in claim 11, and the same rejection is incorporated herein. Poulis discloses generate the first input including the input prompt, the domain-specific key-value dataset, and the domain-specific ontology map Figure 4, item 404; paragraph 0053).
Poulis fails to specifically disclose:
obtain a simulated dataset associated with a data transformation pipeline comprising a simulated node data for the one or more nodes of the node set
wherein the simulated dataset is generated using the domain-specific ontology map
generate the input prompt including a request to validate the simulated dataset against the domain-specific ontology map
generate the first input including the simulated dataset
However, Panwar, which is analogous to the claimed invention because it is directed toward generating simulated data, discloses:
obtain a simulated dataset associated with a data transformation pipeline comprising a simulated node data for the one or more nodes of the node set (Figure 5; paragraphs 0100-0101: Here, synthetic data generation module performs a statistical analysis to determine relevance of epoch nodes)
wherein the simulated dataset is generated using the domain-specific ontology map (Figure 5; paragraphs 0100-0101: Here, synthetic data generation module performs a statistical analysis to determine relevance based on domain-specific data)
generate the input prompt including a request to validate the simulated dataset against the domain-specific ontology map (Figure 5; paragraphs 0100-0101: Here, synthetic data generation module performs a statistical analysis to determine relevance of epoch nodes)
generate the first input including the simulated dataset (Figure 5; paragraphs 0100-0101: Here, synthetic data generation module performs generation of synthetic data)
It would have been obvious to one of ordinary skill in the art at the time of the applicant’s effective filing date to have combined Panwar with Poulis-Minkin, with a reasonable expectation of success, as it would have allowed for generating synthetic data based upon domain-specific applications to improve synthetic datasets (Panwar: paragraphs 0100-0101).
Response to Arguments
Applicant’s arguments with respect to the rejection of claims under 35 USC 101 have been fully considered and are persuasive.
The combination of arguments regarding limitations considered under Step 2A, Prong One (pages 17-19), the arguments regarding an improvement to the functioning of a computer under Step 2A, Prong Two (pages 19-22), and the consideration of limitations individually and as a combination under Step 2B (page 22) is persuasive.
The rejection of claims under 35 USC 101 has been withdrawn.
Applicant’s arguments with respect to the rejection of claims under 35 USC 103 with respect to the amended limitations have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground of rejection is made in view of Poulis, Minkin, and Kashyap.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Katam et al. (US 2026/0046213): Discloses validation rules (paragraph 0009) applied to a prompt template (paragraph 0012)
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KYLE R STORK whose telephone number is (571)272-4130. The examiner can normally be reached 8am - 2pm; 4pm - 6pm.
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, Omar Fernandez Rivas can be reached at 571/272-2589. 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, visit: 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.
/KYLE R STORK/Primary Examiner, Art Unit 2128
Prosecution Insights