DETAILED ACTION
This Office action is in response to Applicant’s reply filed 05/28/2026.
Claims 1-6 and 8-20 are pending.
Claim 7 is canceled.
Claims 1-2, 4-6, 8-11, 13-17, and 19-20 are amended.
Claims 1-6 and 8-20 are rejected.
Notice of 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 .
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on 05/28/2026 and 08/03/2026 were filed prior to this Office action. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement(s) is/are being considered by the examiner.
Examiner Notes/Objections
Claim 1 was objected to but has been corrected. All objections are hereby withdrawn.
Statutory Review under 35 USC § 101
Claims 1-9 are directed toward a system and have been reviewed.
Claims 1-9 initially appear to be statutory, as the system includes hardware (a non-transitory memory).
Claims 1-6 and 8-9 also appear to be patent-eligible at this time as they now recite the elements of claim 7, as a whole integrating the abstract idea into a practical application based on Step 2A, Prong Two of the current patent subject matter eligibility determination.
Claims 10-15 are directed towards a method and have been reviewed.
Claims 10-15 appear to be patent-eligible at this time as they now recite the elements of claim 7, as a whole integrating the abstract idea into a practical application based on Step 2A, Prong Two of the current patent subject matter eligibility determination.
Claims 16-20 are directed toward an article of manufacture and have been reviewed.
Claims initially appear to be statutory, as the article of manufacture excludes transitory signals (claim says non-transitory).
Claims 16-20 also appear to be patent-eligible at this time as they now recite the elements of claim 7, as a whole integrating the abstract idea into a practical application based on Step 2A, Prong Two of the current patent subject matter eligibility determination.
Response to Arguments
Applicant’s arguments, see Remarks pp9-10, filed 05/28/2026, with respect to the 35 U.S.C. 101 rejection of claims 1-6 and 8-20 have been fully considered and are persuasive. The 35 U.S.C. 101 rejection of claims 1-6 and 8-20 has been withdrawn.
Applicant’s arguments, see Remarks p10, filed 05/28/2026, with respect to the rejection(s) of claim(s) 1, 7, 9, 10, and 16 under 35 U.S.C. 102 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III.
Applicant’s arguments, see Remarks pp11-12, filed 05/28/2026, with respect to the rejection(s) of claims 2-6, 8, 11-15, and 17-20 under 35 U.S.C. 103 have been fully considered; however, a new ground(s) of rejection is made of parent independent claims 1, 10, and 16 under 35 U.S.C.103. As a result, claims 2-6, 8, 11-15, and 17-20 remain rejected at least by virtue of their dependence on rejected base claims.
Claim Rejections - 35 USC § 103
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.
Claims 1, 9; 10; and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar et al., U.S. Patent Application Publication No. 2025/0217769 (filed April 11, 2024; hereinafter Kumar) in view of Pack, III et al., U.S. Patent Application Publication No. 2025/0086011 (filed September 7, 2023; hereinafter Pack, III).
Regarding claim 1, Kumar teaches:
A system, comprising: a non-transitory memory; a processor communicatively coupled to the non-transitory memory, wherein the processor is configured to read a set of instructions to: (Kumar FIG. 14, ¶ 0293-0298: a system bus or other communication mechanism 1414 can provide communication between the processing unit 1402, the power source 1412, the memory 1404, the input device(s) 1406, and the output device(s) 1410. The processing unit 1402 can be implemented as any electronic device capable of processing, receiving, or transmitting data or instructions)
receive an attribute extraction request identifying item element data… (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input; Kumar FIG. 6, ¶ 0238: the generative interface panel 620 includes an input region 630 configured to receive natural language user input 632 and other user input. Other user input may include linked content, non-textual or graphical content, selectable graphical objects, and other elements or objects; Kumar FIG. 7, ¶ 0254: a user may provide an additional natural language user input 732 to the user input region 730. Here, the user input includes the phrase “what is assigned to John?”)
generate each of a first generative prompt and a second generative prompt based on the attribute extraction request and the item element data, (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424. The platform frontend 424 passes the input to a prompt management service 426 that formalizes a prompt suitable for input to a generative output engine 428; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input, and constructs a prompt therefrom at operation 502; Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list)
wherein each of the first generative prompt and the second generative prompt define a data structure of the item element data and one or more attributes to be extracted from the item element data; (Kumar FIG. 13B, ¶ 0288: The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; see Kumar ¶ 0184 showing a defined data structure and also first and second prompts: the predetermined prompt text may include example input-output pairs that mirror a desired response style or format for use with the documentation platform ... the configuration module 254 also includes a designation of one or more large-language modules or other resources that can be used by the prompt management service 220 to service the constructed prompts or queries)
configure a first generative model based on the first generative prompt to extract a first value of the one or more attributes identified in the attribute extraction request; (Kumar FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name"; ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine; see relevant ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200)
configure a second generative model based on the second generative prompt to extract a second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments)
extract, by the first generative model, the first value of the one or more attributes; (Kumar ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
extract, by the second generative model, the second value of the one or more attributes; (Kumar shows use of a second generative model ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see also relevant ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments; Kumar shows extraction in ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; and in FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
generate a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0273: the natural language user input is divided into multiple sub-parts or portions, each portion used to generate a separate prompt. The respective results from the prompts can then be recombined or formulated to generate a complete structured query that is executed with respect to the issue tracking platform)
implement an attribute-based automated process based on at least one attribute value in the final attribute set; and (Kumar FIG. 4B, ¶ 0220: Output of the request queue 438 can be provided as input to a prompt hydrator 440 configured to populate template fields, add context identifiers, supplement the prompt, and perform other normalization operations described herein)
store, in a data storage medium, the final attribute set as attribute values for the item element data in accordance with the data structure of the item element data. (Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine [relevant to data structure of the item element data]; see this in light of the storage of Kumar ¶ 0186, "successful outcomes (indicated either expressly through user feedback or implicitly through use of the results) may be stored in the persistence module(s) 224 and used to direct subsequent queries or operations to a similar positive result," ¶ 0193, "the correlation may be based on previous inputs or results stored in the persistence module 224 or otherwise accessible to the system 200," and ¶ 0197, "the prompt management service 220 also has access to or includes a persistence module 224, which may include a cache or other storage of previous user inputs, generative outputs, or other interactions with the content discovery and generation interface 210")
Kumar does not expressly disclose an attribute extraction request identifying one or more attributes.
An interpretation of Kumar results in it not expressly disclosing to generate a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes…
However, Pack, III addresses this by teaching the following:
receive an attribute extraction request identifying item element data and one or more attributes; (Pack, III ¶ 0011-0012: Using any suitable input device, a user may input a prompt to a computing device ... The request may be a job to be accomplished, for example, a request for the generation of computer code for a website, application, web application, or website component that includes certain features specified in the prompt, a request for the generation of data of a specific type, such as product descriptions for product in a database, or a request for the generation of any other suitable data type. The prompt may be phrased in any suitable manner; ¶ 0029: The input data for this composable asynchronous task may include product attributes such as product name, category, features, and pricing information. The desired output format for output from the composable asynchronous task may be, for example, products descriptions for each product that follows a consistent structure, such as, for example, starting with a headline followed by a brief overview of the product, and then a detailed list of features and benefits of the product; FIG. 5, ¶ 0049: At 502, a prompt input by a user may be received)
generate a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes… (Pack, III FIG. 5, ¶ 0051: At 506, the composable asynchronous tasks may be performed, including generating output with an LLM, validating the output of the LLM, and generating output using the output from the LLM … The composable asynchronous tasks 188 may generate output that incorporate the output of the LLM used by the composable asynchronous task 182, as output by the composable asynchronous tasks 182, for example, combining the output of the composable asynchronous tasks 182 with outputs from the composable asynchronous tasks 184 and 186 to generate an output that is responsive to the request in the prompt received by the agent LLM 110)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the prompt generation of Kumar with the prompt execution of Pack, III.
In addition, both of the references (Kumar and Pack, III) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as prompt generation and execution.
Motivation to do so would be to improve the functioning of Kumar processing input prompts with the ability in similar reference Pack, III also processing input prompts generating query prompts but with the improvement of scalability and parallel processing.
Regarding claim 10, Kumar teaches:
A computer-implemented method, comprising: receiving an attribute extraction request identifying item element data… (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input; Kumar FIG. 6, ¶ 0238: the generative interface panel 620 includes an input region 630 configured to receive natural language user input 632 and other user input. Other user input may include linked content, non-textual or graphical content, selectable graphical objects, and other elements or objects; Kumar FIG. 7, ¶ 0254: a user may provide an additional natural language user input 732 to the user input region 730. Here, the user input includes the phrase “what is assigned to John?”)
generating each of a first generative prompt and a second generative prompt based on the attribute extraction request and the item element data, (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424. The platform frontend 424 passes the input to a prompt management service 426 that formalizes a prompt suitable for input to a generative output engine 428; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input, and constructs a prompt therefrom at operation 502; Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list)
wherein each of the first generative prompt and the second generative prompt define a data structure of the item element data and one or more attributes to be extracted from the item element data; (Kumar FIG. 13B, ¶ 0288: The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; see Kumar ¶ 0184 showing a defined data structure and also first and second prompts: the predetermined prompt text may include example input-output pairs that mirror a desired response style or format for use with the documentation platform ... the configuration module 254 also includes a designation of one or more large-language modules or other resources that can be used by the prompt management service 220 to service the constructed prompts or queries)
configuring a first generative model based on the first generative prompt to extract a first value of the one or more attributes identified in the attribute extraction request; (Kumar FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name"; ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine; see relevant ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200)
configuring a second generative model based on the second generative prompt to extract a second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments)
extracting, by the first generative model, the first value of the one or more attributes; (Kumar ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
extracting, by the second generative model, the second value of the one or more attributes; (Kumar shows use of a second generative model ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see also relevant ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments; Kumar shows extraction in ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; and in FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0273: the natural language user input is divided into multiple sub-parts or portions, each portion used to generate a separate prompt. The respective results from the prompts can then be recombined or formulated to generate a complete structured query that is executed with respect to the issue tracking platform)
implementing an attribute-based automated process based on at least one attribute value in the final attribute set; and (Kumar FIG. 4B, ¶ 0220: Output of the request queue 438 can be provided as input to a prompt hydrator 440 configured to populate template fields, add context identifiers, supplement the prompt, and perform other normalization operations described herein)
storing, in a data storage medium, the final attribute set as attribute values for the item element data in accordance with the data structure of the item element data. (Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine [relevant to data structure of the item element data]; see this in light of the storage of Kumar ¶ 0186, "successful outcomes (indicated either expressly through user feedback or implicitly through use of the results) may be stored in the persistence module(s) 224 and used to direct subsequent queries or operations to a similar positive result," ¶ 0193, "the correlation may be based on previous inputs or results stored in the persistence module 224 or otherwise accessible to the system 200," and ¶ 0197, "the prompt management service 220 also has access to or includes a persistence module 224, which may include a cache or other storage of previous user inputs, generative outputs, or other interactions with the content discovery and generation interface 210")
Kumar does not expressly disclose an attribute extraction request identifying one or more attributes.
An interpretation of Kumar results in it not expressly disclosing generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes…
However, Pack, III addresses this by teaching the following:
receiving an attribute extraction request identifying item element data and one or more attributes; (Pack, III ¶ 0011-0012: Using any suitable input device, a user may input a prompt to a computing device ... The request may be a job to be accomplished, for example, a request for the generation of computer code for a website, application, web application, or website component that includes certain features specified in the prompt, a request for the generation of data of a specific type, such as product descriptions for product in a database, or a request for the generation of any other suitable data type. The prompt may be phrased in any suitable manner; ¶ 0029: The input data for this composable asynchronous task may include product attributes such as product name, category, features, and pricing information. The desired output format for output from the composable asynchronous task may be, for example, products descriptions for each product that follows a consistent structure, such as, for example, starting with a headline followed by a brief overview of the product, and then a detailed list of features and benefits of the product; FIG. 5, ¶ 0049: At 502, a prompt input by a user may be received)
generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes… (Pack, III FIG. 5, ¶ 0051: At 506, the composable asynchronous tasks may be performed, including generating output with an LLM, validating the output of the LLM, and generating output using the output from the LLM … The composable asynchronous tasks 188 may generate output that incorporate the output of the LLM used by the composable asynchronous task 182, as output by the composable asynchronous tasks 182, for example, combining the output of the composable asynchronous tasks 182 with outputs from the composable asynchronous tasks 184 and 186 to generate an output that is responsive to the request in the prompt received by the agent LLM 110)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the prompt generation of Kumar with the prompt execution of Pack, III.
In addition, both of the references (Kumar and Pack, III) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as prompt generation and execution.
Motivation to do so would be to improve the functioning of Kumar processing input prompts with the ability in similar reference Pack, III also processing input prompts generating query prompts but with the improvement of scalability and parallel processing.
Regarding claim 16, Kumar teaches:
A non-transitory computer readable medium having instructions stored thereon, wherein the instructions, when executed by at least one processor, cause the at least one device to perform operations comprising: (Kumar FIG. 14, ¶ 0293-0298: a system bus or other communication mechanism 1414 can provide communication between the processing unit 1402, the power source 1412, the memory 1404, the input device(s) 1406, and the output device(s) 1410. The processing unit 1402 can be implemented as any electronic device capable of processing, receiving, or transmitting data or instructions)
receiving an attribute extraction request identifying item element data… (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input; Kumar FIG. 6, ¶ 0238: the generative interface panel 620 includes an input region 630 configured to receive natural language user input 632 and other user input. Other user input may include linked content, non-textual or graphical content, selectable graphical objects, and other elements or objects; Kumar FIG. 7, ¶ 0254: a user may provide an additional natural language user input 732 to the user input region 730. Here, the user input includes the phrase “what is assigned to John?”)
generating each of a first generative prompt and a second generative prompt based on the attribute extraction request and the item element data, (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424. The platform frontend 424 passes the input to a prompt management service 426 that formalizes a prompt suitable for input to a generative output engine 428; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input, and constructs a prompt therefrom at operation 502; Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list)
wherein each of the first generative prompt and the second generative prompt define a data structure of the item element data and one or more attributes to be extracted from the item element data; (Kumar FIG. 13B, ¶ 0288: The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine. In this example, the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; see Kumar ¶ 0184 showing a defined data structure and also first and second prompts: the predetermined prompt text may include example input-output pairs that mirror a desired response style or format for use with the documentation platform ... the configuration module 254 also includes a designation of one or more large-language modules or other resources that can be used by the prompt management service 220 to service the constructed prompts or queries)
configuring a first generative model based on the first generative prompt to extract a first value of the one or more attributes identified in the attribute extraction request; (Kumar FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name"; ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine; see relevant ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200)
configuring a second generative model based on the second generative prompt to extract a second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments)
extracting, by the first generative model, the first value of the one or more attributes; (Kumar ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
extracting, by the second generative model, the second value of the one or more attributes; (Kumar shows use of a second generative model ¶ 0198: the prompt management service 220 is used to create text-based prompts that are transmitted to a generative output engine 230, which may include a LLM or other generative service ... the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200; see also relevant ¶ 0168: the generative output service 116 can be called more than once (and/or it may be configured to generate its own follow-up prompts or prompt templates which can be populated with appropriate information and re-submitted to the generative output service 116 to obtain further generative output. More simply, in some embodiments, generative output may be recursive, iterative, or otherwise multi-step in some embodiments; Kumar shows extraction in ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... the schema example 1384 specifies a proposed structure for providing a set of multiple task lists and associated card topics for each respective task list; and in FIG. 13B: The format of your response should be a valid JSON document that looks like... [{"list_name": "List 1 name", "cards": ["Card 1 name", "Card 2 name", "Card 3 name")
generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes identified in the attribute extraction request; (Kumar ¶ 0273: the natural language user input is divided into multiple sub-parts or portions, each portion used to generate a separate prompt. The respective results from the prompts can then be recombined or formulated to generate a complete structured query that is executed with respect to the issue tracking platform)
implementing an attribute-based automated process based on at least one attribute value in the final attribute set; and (Kumar FIG. 4B, ¶ 0220: Output of the request queue 438 can be provided as input to a prompt hydrator 440 configured to populate template fields, add context identifiers, supplement the prompt, and perform other normalization operations described herein)
storing, in a data storage medium, the final attribute set as attribute values for the item element data in accordance with the data structure of the item element data. (Kumar FIG. 13B, ¶ 0288: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine ... The prompt 1380 also includes an example text-based schema 1384 defining the preferred schema for the output or response of the generative output engine [relevant to data structure of the item element data]; see this in light of the storage of Kumar ¶ 0186, "successful outcomes (indicated either expressly through user feedback or implicitly through use of the results) may be stored in the persistence module(s) 224 and used to direct subsequent queries or operations to a similar positive result," ¶ 0193, "the correlation may be based on previous inputs or results stored in the persistence module 224 or otherwise accessible to the system 200," and ¶ 0197, "the prompt management service 220 also has access to or includes a persistence module 224, which may include a cache or other storage of previous user inputs, generative outputs, or other interactions with the content discovery and generation interface 210")
Kumar does not expressly disclose an attribute extraction request identifying one or more attributes.
An interpretation of Kumar results in it not expressly disclosing generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes…
However, Pack, III addresses this by teaching the following:
receiving an attribute extraction request identifying item element data and one or more attributes; (Pack, III ¶ 0011-0012: Using any suitable input device, a user may input a prompt to a computing device ... The request may be a job to be accomplished, for example, a request for the generation of computer code for a website, application, web application, or website component that includes certain features specified in the prompt, a request for the generation of data of a specific type, such as product descriptions for product in a database, or a request for the generation of any other suitable data type. The prompt may be phrased in any suitable manner; ¶ 0029: The input data for this composable asynchronous task may include product attributes such as product name, category, features, and pricing information. The desired output format for output from the composable asynchronous task may be, for example, products descriptions for each product that follows a consistent structure, such as, for example, starting with a headline followed by a brief overview of the product, and then a detailed list of features and benefits of the product; FIG. 5, ¶ 0049: At 502, a prompt input by a user may be received)
generating a final attribute set including at least a portion of the first value and at least a portion of the second value of the one or more attributes… (Pack, III FIG. 5, ¶ 0051: At 506, the composable asynchronous tasks may be performed, including generating output with an LLM, validating the output of the LLM, and generating output using the output from the LLM … The composable asynchronous tasks 188 may generate output that incorporate the output of the LLM used by the composable asynchronous task 182, as output by the composable asynchronous tasks 182, for example, combining the output of the composable asynchronous tasks 182 with outputs from the composable asynchronous tasks 184 and 186 to generate an output that is responsive to the request in the prompt received by the agent LLM 110)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the prompt generation of Kumar with the prompt execution of Pack, III.
In addition, both of the references (Kumar and Pack, III) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as prompt generation and execution.
Motivation to do so would be to improve the functioning of Kumar processing input prompts with the ability in similar reference Pack, III also processing input prompts generating query prompts but with the improvement of scalability and parallel processing.
Regarding claim 9, Kumar in view of Pack, III teaches:
wherein the processor is configured to read the instructions to cause a display of the at least one attribute value in the final attribute set in conjunction with interface elements associated with the item element data. (Kumar FIGs. 13A-13B, ¶ 0288-0289: the prompt 1380 may include a query or request portion 1382, which provides the overall context for the request and general parameters of what should be returned by the generative output engine; Kumar FIG. 9, ¶ 0267-0268: In response, the generative output engine produces a generative output or response that includes a proposed structured query having a format consistent with the schema compatible to the particular issue tracking platform; Kumar FIG. 6, ¶ 0245: The generative response may then be used to produce the response 642 displayed in the generative interface panel 620. As described previously, the system may perform post-processing on the generative response in order to substitute elements of the response with system objects or with identifiers that map to the elements of the response. In this case, the post-processing inserts issue identifier objects 642, which may reference or link to issues managed by an issue tracking system)
Claims 2, 11, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III in view of Nguyen et al., U.S. Patent Application Publication No. 2020/0301916 (hereinafter Nguyen).
Regarding claims 2, 11, and 17, Kumar in view of Pack, III teaches all the features with respect to claims 1, 10, and 16 above respectively including:
generate the first and second generative prompts… (Kumar FIG. 4B, ¶ 0218: a user input 422 may be received at a platform frontend 424. The platform frontend 424 passes the input to a prompt management service 426 that formalizes a prompt suitable for input to a generative output engine 428; Kumar FIG. 5A, ¶ 0223: The system 500a receives user input, and constructs a prompt therefrom at operation 50)
Kumar in view of Pack, III does not expressly disclose:
determine a classifier associated with the item element data;
receive an attribute extraction template based on the determination;
Kumar in view of Pack, III further does not expressly disclose the prompt being based on the attribute extraction template.
However, Nguyen addresses this by teaching:
determine a classifier associated with the item element data; (Nguyen ¶ 0032: each set of users interacting with a system can customize the system to process natural language queries typically asked in a particular domain; Nguyen ¶ 0132: The ability to add query templates allows organizations/enterprises to build their own database of query templates that is capable of processing the typical queries that users perform in that domain ... The queries based on a query intent take as input the set of attributes associated with the query intent. For example, the query intent of comparing two attributes takes as input at least a first attribute and a second attribute. Each of these query templates specifies the same set of attributes, i.e., the set of attributes associated with the query intent)
receive an attribute extraction template based on the determination; and (Nguyen FIG. 7, ¶ 0108-0109: The suggestion module 370 matches 720 the input query string against templates of natural language queries stored in the query template store 390. The suggestion module 370 identifies terms of the input query string and matches the terms of query templates in the order in which the terms occur in the natural language query and the order in which the query template expects the terms ... the query template may specify that a term can be an attribute or a user defined metric)
generate the … prompts based on the attribute extraction template. (Nguyen FIG. 7, ¶ 0109-0115: The suggestion module 370 matches 720 the input query string against templates of natural language queries stored in the query template store 390 … The user interaction module 240 determines the terms to be suggested in response to the received query string)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the prompt generation of Kumar as modified with the prompt generation of Nguyen.
In addition, both of the references (Kumar as modified and Nguyen) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as prompt generation and presentation.
Motivation to do so would be to improve the functioning of Kumar as modified generating query prompts with the ability in similar reference Nguyen also generating query prompts but with the improvement of query template comparisons.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to provide suitable interface for users to analyze the large amount of information available in an enterprise as seen in Nguyen ¶ 0004.
Claims 3, 5; 12, 14; 18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III in view of Guerra et al., U.S. Patent No. 12,346,314 (filed July 16, 2024, prior to the instant application date of July 30, 2024; hereinafter Guerra).
Regarding claims 3, 12, and 18, Kumar in view of Pack, III teaches all the features with respect to claims 1, 10, and 16 above respectively including:
receive attribute model configuration data, and determine the one or more attributes to be extracted from the … data based on the attribute model configuration data. (Kumar FIG. 9, ¶ 0267: The prompt 900 also includes a set of structured query examples 908 that provide demonstrative input-output pairs. Specifically, the input-output pairs include an example natural language input or prompt paired with an example schema-formatted output; Kumar FIG. 13B, ¶ 0288: The prompt 1380 also includes other rules and instructions 1386 that specify permitted results and prohibited results in order to further guide the generative output engine. Similar to other previous examples, prompt 1380 may also include example input-output pairs that provide further guidance for expected output in response to specified example input)
Kumar in view of Pack, III does not expressly disclose the one or more attributes to be extracted from the item element data.
However, Guerra addresses this by teaching one or more attributes extracted from the item element data. (Guerra FIG. 4, step 422, col. 11, line 56-col. 12, line 16: a table column may have a cryptic field name (e.g., “OSDSTATUS”) and/or may contain data values with specific notations (e.g., categories “A,” “B,” “C,” or space). If such data is sent directly to a generative AI model, it may not be able to interpret the data correctly. However, by replacing these cryptic field names and data values with descriptive metadata (e.g., replacing “OSDSTATUS” with “Sales Order Overall Processing Status,” and replacing categorical data value “A” with “Complete,” or the like), the data becomes more meaningful and interpretable for the generative AI model to generate more contextually relevant responses)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the generative query techniques of Kumar as modified with the generative query techniques of Guerra.
In addition, both of the references (Kumar as modified and Guerra) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as generative query techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to provide more meaningful and interpretable data for a generative model to generate more contextually relevant responses as seen in Guerra col. 12, lines 9-16.
Regarding claims 5, 14, and 20, Kumar in view of Pack, III teaches all the features with respect to claims 1, 10, and 16 above respectively but does not expressly disclose:
determine at least a portion of the first and second generative prompts based on an associated type of attribute.
However, Guerra addresses this by teaching:
determine at least a portion of the first and second generative prompts based on an associated type of attribute. (Guerra FIG. 4, steps 422-424, col. 11, line 56-col. 12, line 38: by replacing these cryptic field names and data values with descriptive metadata (e.g., replacing “OSDSTATUS” with “Sales Order Overall Processing Status,” and replacing categorical data value “A” with “Complete,” or the like), the data becomes more meaningful and interpretable for the generative AI model to generate more contextually relevant responses ... Based on the intent of the user query, the method can select the prompt template from a plurality of prompt templates, each corresponding to a specific operation mode ... For an operation mode to obtain processing status of a specific sales order object, the prompt template can include instructions for the generative AI model to generate a list of one or more business objects (including the sales order object) and their status information in a sequential order)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the generative query techniques of Kumar as modified with the generative query techniques of Guerra.
In addition, both of the references (Kumar as modified and Guerra) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as generative query techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to provide more meaningful and interpretable data for a generative model to generate more contextually relevant responses as seen in Guerra col. 12, lines 9-16.
Claims 4, 13, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III in view of Bharadwaj et al., U.S. Patent Application Publication No. 2024/0273345 (filed February 13, 2023, prior to the instant application date of July 30, 2024; hereinafter Bharadwaj).
Regarding claims 4, 13, and 19, Kumar in view of Pack, III teaches all the features with respect to claims 1, 10, and 16 above respectively including:
generate each of the first and second generative prompts to comprise one or more configurations for the first and second generative models, and configure the first and second generative models… (Kumar ¶ 0198: the generative output engine 230 may one of a variety of generative output engines or models that are associated with the system 200. In some implementations, a different model may be associated with different assistant services and may be selected or adapted for use with a particular subject-matter expertise of a respective assistant service. For example, a model having a corpus or set of tokens that are tailored for a particular subject matter or type of inquiry may be selected from a set of models in order to provide generative content for a respective assistant service tailored for a similar subject matter or type of inquiry)
Kumar in view of Pack, III does not expressly disclose to configure based on the one or more configurations.
However, Bharadwaj addresses this by teaching the following:
generate each of the first and second generative prompts to comprise one or more configurations for the first and second generative models, and (see prompts in Bharadwaj FIG. 1A, ¶ 0114-0116: The one or more characteristics of the input query may include, for instance, a user identifier and a user prompt ... after determining one or more characteristics of the input query at step 104, the method 100 may proceed to step 106a, wherein step 106a comprises selecting, from a plurality of response modules, one or more response modules based on the one or more characteristics of the input query and one or more user preference metrics associated with each of the one or more response modules; see configurations in Bharadwaj ¶ 0126-0129: each model forming a response module further comprises configurations/configuration files for optimizing one or more parameters, settings, preferences, etc. of each respective machine learning model of the plurality of machine learning models)
configure the first and second generative models based on the one or more configurations. (Bharadwaj FIG. 1A, ¶ 0114-0116: step 106a comprises selecting, from a plurality of response modules, one or more response modules based on the one or more characteristics of the input query and one or more user preference metrics associated with each of the one or more response modules; see then Bharadwaj ¶ 0126-0129: each model forming a response module further comprises configurations/configuration files for optimizing one or more parameters, settings, preferences, etc. of each respective machine learning model of the plurality of machine learning models)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the generative query techniques of Kumar as modified with the generative query techniques of Bharadwaj.
In addition, both of the references (Kumar as modified and Bharadwaj) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as generative query techniques.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to improve various prompts based on characteristics of a user that provided an input query as seen in Bharadwaj ¶ 0123.
Claims 6 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III in view of Burton, U.S. Patent No. 12,210,839 (filed April 29, 2024, prior to the instant application date of July 30, 2024; hereinafter Burton).
Regarding claims 6 and 15, Kumar in view of Pack, III teaches all the features with respect to claims 1 and 10 above respectively but does not expressly disclose:
generate the first and second generative prompts to comprise a plurality of attribute definitions for each of the one or more attributes, and configure the first and second generative models based on the plurality of attribute definitions.
However, Burton addresses this by teaching:
generate the first and second generative prompts to comprise a plurality of attribute definitions for each of the one or more attributes, and (Burton col. 57, lines 18-44: the neural networks performing framing analysis inference with event-related tags and types may be arranged in any economic configuration, including but not limited to: omnibus classifiers; one vs. rest ensembles with a consensus step; or few-shot methods which use agentic or standard completion LLMs to answer written prompts asking questions about framing attributes directly (e.g. “Considering the definitions of channels I have provided to you, please classify the following sentence with one of the provided channel names: <sentence>.”) ... the tag determination may be done by prompting a GPT-type model with information related to the channel (e.g. “Given this list of tags concerning Revenue themes: <list of tags>, which 0-3 do you consider to be present in the following text: <text>?”))
configure the first and second generative models based on the plurality of attribute definitions. (Burton col. 57, lines 18-44: The models may be trained by the use of the training module which performs a reverse operation to the inferencer, which consumes hierarchically organized annotation datum files and performing inference on the text extracted from and referenceable to any document markup (e.g. HTML) which may contain the text in the original textual material)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the generative models of Kumar as modified with the generative models of Burton.
In addition, both of the references (Kumar as modified and Burton) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as generative model prompts.
Motivation to do so would be to fortify the teachings of Kumar as modified involving training generative models with similar reference Burton also training generative models.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to allow quality of results as experienced within the interactive portion of the system to improve over time as seen in Burton col. 57, lines 55-62.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Kumar in view of Pack, III in view of Cheah et al., U.S. Patent Application Publication No. 2023/0418853 (hereinafter Cheah).
Regarding claim 8, Kumar in view of Pack, III teaches all the features with respect to claim 1 above but does not expressly disclose:
extract, by the first and second generative models, a confidence value associated with each of the one or more attributes, and
generate the final attribute set to include, for each of the one or more attributes, the value associated with a highest corresponding confidence value.
However, Cheah addresses this by teaching:
extract, by the first and second generative models, a confidence value associated with each of the one or more attributes, and (Cheah FIG. 3, ¶ 0068-0070: The attribute prediction model 236 may be a generative model, that is, one that approximates the probability of an output Y for a candidate token based on the preliminary labels of the candidate token without needing ground truth Y ... Based on the confidence levels for the attribute label(s) (e.g. ABN, BSB and ACC) for the tokens of the subset, the document labelling module 238 may select suitable attribute label values for the document. For example, the document labelling module 238 may be configured to determine a token having a highest confidence value for a specific attribute label as being the value for that specific attribute label for the document)
generate the final attribute set to include, for each of the one or more attributes, the value associated with a highest corresponding confidence value. (Cheah FIG. 3, ¶ 0075-0076, "At 306, the system 202 determines a set of preliminary attribute labels for each of the one or more tokens" and ¶ 0091, "At 312, the system 202 determines a set of refined labels for each document based on the confidence values associated with the tokens of the respective subset of tokens. The set of refined labels comprises a value for the one or more attribute types"; see this in light of Cheah ¶ 0070: the document labelling module 238 may be configured to determine a token having a highest confidence value for a specific attribute label as being the value for that specific attribute label for the document)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the functioning of the data analysis of Kumar as modified with the attribute analysis of Cheah.
In addition, both of the references (Kumar as modified and Cheah) disclose features that are directed to analogous art, and they are directed to the same field of endeavor, such as data derivation techniques.
Motivation to do so would be to improve the functioning of Kumar as modified identifying data with the ability in similar reference Cheah also identifying data but with the improvement of confidence values.
Motivation to do so would also be the teaching, suggestion, or motivation for a person of ordinary skill in the art to provide efficient and accurate extraction of relevant data as seen in Cheah ¶ 0003.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Shetty et al., U.S. Patent Application Publication No. 2026/0030530; see Shetty ¶ 0037 describing a prompt engineered to include an output format directive, ¶ 0049 describing an output format directive and providing an example format for a table or a link to a desired template being provided in the prompt, and ¶ 0053 describing more examples of desired output, relevant to at least the independent claims involving the prompts defining a data structure and attributes to be extracted.
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 JEDIDIAH P FERRER whose telephone number is (571)270-7695. The examiner can normally be reached Monday, Tuesday, Friday, 12:00pm-9:00pm.
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, Kavita Stanley can be reached at (571)272-8352. 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.
/J.P.F/Examiner, Art Unit 2153 August 22, 2026
/KAVITA STANLEY/Supervisory Patent Examiner, Art Unit 2153
Prosecution Insights