DETAILED ACTION
This office action is in response to the communication filed on July 08, 2026. Claims 1-11 and 13-21 are currently pending.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 07/08/26 has been entered.
Response to Arguments
Applicant's arguments filed on July 08, 2026 have been fully considered but they are not persuasive for the following reasons:
Applicant in Pages 10-17 of the Remarks argues that even if applicant’s amended claims are directed to a judicial exception, the claims amount to significantly more than the judicial exception because the claims provide an improvement to computer functionality, and that the amended claims explicitly recite the vectorized databases disclosed in applicant’s specification that result in computational improvements.
Applicant further argues that the amended independent claims 1, 19, and 20 and the amended dependent claims 16, 18, and 21 recite vectorized databases providing different levels of granularity that may help the search process when executing a search query, which clearly provides an improvement to computer functionality by improving the query search process.
Examiner respectfully disagrees.
It is important to note that the judicial exception alone cannot provide the
improvement. The improvement can be provided by one or more additional elements
(MPEP 2106.05(a)).
If the specification sets forth an improvement in technology, the claim must be evaluated to ensure that the claim itself reflects the disclosed improvement (MPEP 2106.04(d)(1)).
Examiner notes that the argued computational improvements disclosed in applicant’s specification, such as providing different levels of granularity that may help the search process when executing a search query, are either not reflected in the recited claims or are argued to be reflected in claim limitations that recite an abstract idea.
Independent claim 1 and similarly independent claims 19 and 20 covers several steps, such as the generating steps, that recite an abstract idea within the “Mental Processes” grouping of abstract ideas, because a person can mentally or using a pen and paper perform the limitations recited in said steps, which is discussed in detail in the current 101 rejection below.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
The remaining steps in the claims that are identified as reciting additional elements, such as the constructing, receiving, obtaining, generating, and transmitting steps in claims 1, 19, and 20, are only adding insignificant extra-solution activity to the judicial exception, are recognized as a well understood, routine, and conventional activity within the field of computer functions, and are applying the exception using generic computer components, which is not sufficient to amount to significantly more than the judicial exception and are not directed to any specific improvement in computer technology.
Additionally, dependent claims 2, 6-9, 16, 18, and 21 cover additional steps that also recite an abstract idea within the “Mental Processes” grouping of abstract ideas, as discussed in detail in the 101 rejection below.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Applicant in Pages 17-19 of the Remarks argues that the cited prior art Amatriain-Rubio and McCurdy do not teach or even suggest the features “receiving a plurality of vectorized databases generated according to a process comprising: receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications”, as recited in amended independent claim 1 and similarly recited in amended independent claims 19 and 20.
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., “receiving a plurality of vectorized databases generated according to a process comprising: receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications”) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Examiner is interpreting that applicant meant to argue that the cited prior art Amatriain-Rubio and McCurdy do not teach or even suggest the amended features “constructing a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications”, as recited in amended independent claim 1 and similarly recited in amended independent claims 19 and 20.
Examiner respectfully disagrees. The previously cited prior art Amatriain-Rubio and McCurdy alone and/or in combination discloses the argued features.
In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Amatriain-Rubio in [0066] and [0202] discloses searching for and retrieving information by executing queries on databases, providing retrieved data from databases.
Amatriain-Rubio in [0209] and [0210] discloses data stores implemented using databases, data written to and read from data stores, databases organize and store data records.
Therefore, Amatriain-Rubio discloses constructing a plurality of databases.
Amatriain-Rubio does not explicitly disclose:
…a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications;
McCurdy in [0005] and [0006] discloses a vector database comprising a plurality of vectors, each vector of the plurality of vectors representative of a text segment of a plurality of text segments, database configured to store vector embeddings representative of text segments.
McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data.
McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database.
McCurdy in [0027]-[0029] discloses vector databases store vector information representative of natural-language text, vectors stored in vector database represent any suitable length of text, such as sentences, paragraphs, etc., storing vector embeddings useful for generating relevant responses, databases storing application data, the vectors representing sentences within messages and/entire messages sent through chat, and vector database can represent technical support chat histories, represent both user queries and responses, collection of chat history of messages including prompts and responses, vector database can separate chat history into messages and/or sentences and store vector embeddings of those text segments in vector database, which is summaries of features of chats and response related to technical support discussions.
McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions.
McCurdy in [0046] and [0051] discloses vectors of vector database include embedding dimensions that correspond to the types of retrievable data, using similarity in vector embedding dimensions of query vector and database vectors to narrow search space, accurately identifying relevant information, reducing the likelihood that irrelevant information is retrieved.
Therefore, McCurdy discloses a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications.
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, having the teachings of Amatriain-Rubio and McCurdy, to have combined Amatriain-Rubio and McCurdy. The motivation to combine Amatriain-Rubio and McCurdy would be to improve the accuracy and/or relevance of natural-language generated by a model in response to a prompt by augmenting prompts with additional information to provide context to the language model.
For the above reasons, Examiner states that rejection of the current Office action is proper.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-11 and 13-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
At step 1:
Independent claims 1, 19, and 20 respectively recite a method, a computer
system, and a non-transitory machine-readable medium, which are directed to a
statutory category such as a process, machine, or an article of manufacture.
At step 2A, prong one:
Independent claim 1 and similarly independent claims 19 and 20 recite the limitations:
“…generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications”;
A person can mentally or using a pen and paper generate a second text description data comprising descriptions of a plurality of applications, a third text description data comprising descriptions of groupings of the plurality of applications, and a fourth text description data comprising descriptions of sets of the groupings of the plurality of applications.
“generating, by way of a first natural language model and based on the first prompt, a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from the plurality of vectorized databases”;
A person can mentally or using a pen and paper generate, by way of a first natural language model and based on a first prompt, a search query comprising a plurality of vector values relating to the first prompt, that are usable to retrieve vectorized data from a plurality of vectorized.
The limitations, as recited above, are processes that, under their broadest reasonable interpretation, cover steps that can be performed in the human mind or by a human using a pen and paper, but for recitation of generic computer components.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
At step 2A, prong two:
This judicial exception is not integrated into a practical application.
Independent claim 1 and similarly independent claims 19 and 20 recite the limitations:
“constructing a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications…”, which is a step of storing received data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“receiving, from a client device, an input query”, which is a step of receiving data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“obtaining, using a first prompt generator, a first prompt based on the input query”, which is a step of obtaining data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data”, which is a step of obtaining data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“obtaining, using a second prompt generator, a second prompt based on the vectorized data and the input query”, which is a step of obtaining data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“generating, by way of a second natural language model and based on the second prompt, a result of the input query”, which is a step of generating result data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
“transmitting the result of the input query to the client device”, which is a step of transmitting data. The step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
The additional elements “a plurality of vectorized databases”, “a plurality of applications”, “from a client device”, “using a first prompt generator”, “a first natural language model”, “using a second prompt generator”, “a second natural language model”, and “to the client device” in the steps in claim 1 are recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
The additional elements “a computing system comprising: one or more processors; memory; and program instructions, stored in the memory, that, upon execution by the one or more processors, cause the one or more processors to perform operations comprising:”, “a plurality of vectorized databases”, “a plurality of applications”, “from a client device”, “using a first prompt generator”, “a first natural language model”, “using a second prompt generator”, “a second natural language model”, and “to the client device” in the steps in claim 19 are recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
The additional elements “a non-transitory machine-readable medium storing program instructions that, when executed by one or more processors of a computing system, cause the one or more processors to perform operations comprising:”, “a plurality of vectorized databases”, “a plurality of applications”, “from a client device”, “using a first prompt generator”, “a first natural language model”, “using a second prompt generator”, “a second natural language model”, and “to the client device” in the steps in claim 20 are recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
At step 2B:
Independent claims 1, 19, and 20 recite the same additional elements as identified in step 2A prong two above. These additional elements are not sufficient to amount to significantly more than the judicial exception.
Independent claim 1 and similarly independent claims 19 and 20 recite the limitations:
“constructing a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications…”, which is a step of storing received data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“receiving, from a client device, an input query”, which is a step of receiving data. The step is recited at a high level of generality, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of receiving or transmitting data over a network (MPEP 2106.05(d)(II)(i)).
“obtaining, using a first prompt generator, a first prompt based on the input query”, which is a step of obtaining data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data”, which is a step of obtaining data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“obtaining, using a second prompt generator, a second prompt based on the vectorized data and the input query”, which is a step of obtaining data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“generating, by way of a second natural language model and based on the second prompt, a result of the input query”, which is a step of generating result data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“transmitting the result of the input query to the client device”, which is a step of transmitting data, and is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of receiving or transmitting data over a network (MPEP 2106.05(d)(II)(i)).
Accordingly, the additional limitations are not sufficient to amount to significantly more than the judicial exception. Therefore, the claims are directed to an abstract idea and are not patent eligible.
Dependent claim 2 recites additional limitations, such as:
“wherein each vectorized database of the plurality of vectorized databases comprises vectorized portions of text”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a search query usable to retrieve vectorized data from vectorized databases comprising vectorized portions of text, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
The additional element “each vectorized database of the plurality of vectorized databases” is recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 3 recites additional limitations, such as:
“wherein the first prompt is obtained by the first prompt generator based on application reference data and context data relating to a user associated with the input query”, which is a step of obtaining data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 4 recites additional limitations, such as:
“wherein the application reference data comprises a list of applications related to a software platform”, which is a step of obtaining data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
The additional element “applications related to a software platform” is recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 5 recites additional limitations, such as:
“wherein the context data relating to the user associated with the input query comprises a subset of the list of applications related to the software platform that have previously been installed by the user associated with the input query”, which is a step of obtaining data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
The additional element “applications related to the software platform that have previously been installed” is recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 6 recites additional limitations, such as:
“wherein the plurality of vector values relating to the first prompt comprises vector values related to each of: the input query, a type classification of the search query, a recommendation, and reduced context data relating to a user associated with the input query”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a search query comprising a plurality of vector values related to each of: an input query, a type classification of the search query, a recommendation, and reduced context data relating to a user associated with the input query, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 7 recites additional limitations, such as:
“wherein the type classification of the search query is determined based on a presence of keywords within the input query”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a search query comprising a plurality of vector values related to a type classification of the search query determined based on a presence of keywords within the input query, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 8 recites additional limitations, such as:
“wherein the recommendation is generated based on a list of applications related to a software platform that have previously been installed by the user associated with the input query”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a search query comprising a plurality of vector values related to a recommendation generated based on a list of applications related to a software platform that have previously been installed by a user associated with an input query, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 9 recites additional limitations, such as:
“wherein the reduced context data is generated by the first natural language model based on context data relating to the user associated with the input query”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a search query comprising a plurality of vector values related to reduced context data relating to a user associated with an input query, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 10 recites additional limitations, such as:
“wherein the second prompt is generated based on reduced context data relating to a user associated with the input query”, which is a step of generating data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 11 recites additional limitations, such as:
“the first natural language model comprises a deep neural network and uses a probabilistic model to generate text; or the first natural language model and the second natural language model are substantially the same”, which is a step further defining the natural language models. The step is recited at a high level of generality, and amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(f)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 13 recites additional limitations, such as:
“executing the search query against a web-based set of documentation related to a software platform to obtain documentation data, wherein the second prompt is obtained using the second prompt generator based on the documentation data”, which is a step of obtaining data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
The additional element “against a web-based set of documentation related to a software platform” is recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 14 recites additional limitations, such as:
“wherein the result of the input query comprises a hyperlink, and wherein the hyperlink refers to additional information related to the result of the input query”, which is a step of generating data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 15 recites additional limitations, such as:
“displaying, on a screen of the client device, the result of the input query”, which is a step of displaying data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering and outputting, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of presenting offers and gathering statistics (MPEP 2106.05(d)(II)(iv)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 16 recites additional limitations, such as:
“generating the second text description data comprises generating…the second text description data”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a second text description data, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“generating the third text description data comprises generating…the third text description data”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a third text description data, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“generating the fourth text description data comprises generating…the fourth text description data”.
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper generate a fourth text description data, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
The additional element “by a third natural language model” and “by the third natural language model” are recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 17 recites additional limitations, such as:
“wherein the third natural language model is configured to generate summaries of input text data”, which is a step further defining the third natural language model. The step is recited at a high level of generality, and amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(f)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 18 recites additional limitations, such as:
wherein constructing the plurality of vectorized database further comprises:
“converting the first text description data into first vectorized portions of text”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper convert a first text description data into first vectorized portions of text, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“storing the first vectorized portions of text in a first vectorized database of the plurality of vectorized databases”, which is a step of storing data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“converting the second text description data into second vectorized portions of text”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper convert a second text description data into second vectorized portions of text, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“storing the second vectorized portions of text in a second vectorized database of the plurality of vectorized databases”, which is a step of storing data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“converting the third text description data into third vectorized portions of text”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper convert a third text description data into third vectorized portions of text, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“storing the third vectorized portions of text in a third vectorized database of the plurality of vectorized databases”, which is a step of storing data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
“converting the fourth text description data into fourth vectorized portions of text”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper convert a fourth text description data into fourth vectorized portions of text, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“storing the fourth vectorized portions of text in a fourth vectorized database of the plurality of vectorized databases”, which is a step of storing data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
The additional element “a first vectorized database of the plurality of vectorized databases”, “a second vectorized database of the plurality of vectorized databases”, “a third vectorized database of the plurality of vectorized databases”, and “a fourth vectorized database of the plurality of vectorized databases” are recited at a high-level of generality, such that it amounts to no more than mere instructions to apply the exception using generic computer components.
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Dependent claim 21 recites additional limitations, such as:
“determining at least one of the plurality of vectorized databases is irrelevant to the search query”;
These limitations are directed to the same abstract idea under the mental processes grouping as independent claim 1, because a person can mentally or using a pen and paper determine that at least one of a plurality of vectorized databases is irrelevant to a search query, and because the limitations do not recite any additional elements that are sufficient to amount to significantly more.
“and upon determining the at least one database is irrelevant to the search query, skipping searching the at least one database when obtaining the vectorized data”, which is a step of searching and obtaining data.
At step 2A prong two, the step is recited at a high level of generality, and amounts to mere data gathering, which is a form of insignificant extra-solution activity (MPEP 2106.05(g)).
At step 2B, the step is recognized as a well understood, routine, and conventional activity within the field of computer functions as an element of storing and retrieving information in memory (MPEP 2106.05(d)(II)(iv)).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claims a whole, because it does not impose any meaningful limits on practicing the abstract idea.
Accordingly, dependent claims 2-11, 13-18, and 21 are also directed to abstract idea without significantly more and are not patent eligible.
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.
Claim(s) 1-11 and 13-21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Amatriain-Rubio (US Pub 2025/0005288) in view of McCurdy (US Pub 2025/0117414).
With respect to claim 1, Amatriain-Rubio discloses a method comprising:
constructing a plurality of…databases… (Amatriain-Rubio in [0066] and [0202] discloses searching for and retrieving information by executing queries on databases, providing retrieved data from databases; Amatriain-Rubio in [0209] and [0210] discloses data stores implemented using databases, data written to and read from data stores, databases organize and store data records; here Amatriain-Rubio does not explicitly discloses a plurality of vectorized databases, but the McCurdy reference discloses the feature, as discussed below);
receiving, from a client device, an input query (Amatriain-Rubio in [0037] and [0054] discloses receiving a user input to a user assistance system from a user device);
obtaining, using a first prompt generator, a first prompt based on the input query (Amatriain-Rubio in [0044] and [0054] discloses user assistance system includes prompt generators and large language models, prompt generator formulates and outputs specific type of prompts that can be used as inputs to one or more large language models, prompt includes instructions to cause a large language model to generate an output in response to the prompt; Amatriain-Rubio in [0054] discloses user;
generating, by way of a first natural language model and based on the first prompt, a search query comprising a plurality of…values relating to the first prompt that are usable to retrieve…data from a plurality of…databases… (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio discloses generating a query using a natural language model and a prompt comprising a plurality of values to search a database, but does not explicitly disclose generating a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases, wherein the plurality of vectorized databases comprises vectorized portions of text relating to (i) summaries of features relating to a plurality of applications and (ii) descriptions of the plurality of applications, but the McCurdy reference discloses the feature, as discussed below);
executing, against the plurality of…databases, the search query using the plurality of…values to obtain the…data (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data, but the McCurdy reference discloses the feature, as discussed below);
obtaining, using a second prompt generator, a second prompt…(Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose prompt based on the vectorized data and the input query, but the McCurdy reference discloses the feature, as discussed below);
generating, by way of a second natural language model and based on the second prompt, a result of the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly discloses generating a result of a prompt based on vectorized data and the input query, but the McCurdy reference discloses the feature; and
transmitting the result of the input query to the client device (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data).
Amatriain-Rubio discloses constructing a plurality of databases, receiving an input query, obtaining a prompt from a prompt generator based on the input query, generating a search query comprising a plurality of values usable to retrieve data from a database using a natural language model and the prompt, executing the query against the database to obtain data, obtaining a second prompt using a second prompt generator, generating a result of an input query and transmitting the result to a client device, however Amatriain-Rubio does not explicitly disclose:
…a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications;
generating…a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from the plurality of vectorized databases;
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data;
obtaining…a second prompt based on the vectorized data and the input query;
generating…based on the second prompt, a result of the input query
transmitting the result of the input query….
The McCurdy reference discloses a plurality of vectorized databases, wherein the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications (McCurdy in [0005] and [0006] discloses a vector database comprising a plurality of vectors, each vector of the plurality of vectors representative of a text segment of a plurality of text segments, database configured to store vector embeddings representative of text segments; McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data; McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database; McCurdy in [0027]-[0029] discloses vector databases store vector information representative of natural-language text, vectors stored in vector database represent any suitable length of text, such as sentences, paragraphs, etc., storing vector embeddings useful for generating relevant responses, databases storing application data, the vectors representing sentences within messages and/entire messages sent through chat, and vector database can represent technical support chat histories, represent both user queries and responses, collection of chat history of messages including prompts and responses, vector database can separate chat history into messages and/or sentences and store vector embeddings of those text segments in vector database, which is summaries of features of chats and response related to technical support discussions; McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions);
generating a search query comprising a plurality of vector values relating to a first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
obtaining a second prompt based on the vectorized data and the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
generating based on the second prompt, a result of the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information); and
transmitting the result of the input query (McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, having the teachings of Amatriain-Rubio and McCurdy, to have combined Amatriain-Rubio and McCurdy. The motivation to combine Amatriain-Rubio and McCurdy would be to improve the accuracy and/or relevance of natural-language generated by a model in response to a prompt by augmenting prompts with additional information to provide context to the language model (McCurdy: [0004]).
With respect to claim 2, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein each vectorized database of the plurality of vectorized databases comprises vectorized portions of text (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model).
With respect to claim 3, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein the first prompt is obtained by the first prompt generator based on application reference data and context data relating to a user associated with the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
With respect to claim 4, Amatriain-Rubio in view of McCurdy discloses the method of claim 3, wherein the application reference data comprises a list of applications related to a software platform (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model).
With respect to claim 5, Amatriain-Rubio in view of McCurdy discloses the method of claim 4, wherein the context data relating to the user associated with the input query comprises a subset of the list of applications related to the software platform that have previously been installed by the user associated with the input query (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model).
With respect to claim 6, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein the plurality of vector values relating to the first prompt comprises vector values related to each of: the input query, a type classification of the search query, a recommendation, and reduced context data relating to a user associated with the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
With respect to claim 7, Amatriain-Rubio in view of McCurdy discloses the method of claim 6, wherein the type classification of the search query is determined based on a presence of keywords within the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
With respect to claim 8, Amatriain-Rubio in view of McCurdy discloses the method of claim 6, wherein the recommendation is generated based on a list of applications related to a software platform that have previously been installed by the user associated with the input query Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model).
With respect to claim 9, Amatriain-Rubio in view of McCurdy discloses the method of claim 6, wherein the reduced context data is generated by the first natural language model based on context data relating to the user associated with the input query (Amatriain-Rubio in [0046] discloses accessing contextual resources and using with one or more language models).
With respect to claim 10, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein the second prompt is generated based on reduced context data relating to a user associated with the input query (Amatriain-Rubio in [052] discloses in response to detecting an increase in latency reducing a number of input signals, or using generative AI models with a reduced size, such as fewer model parameters, or using a more compact template, such as fewer prompt sections, or reducing the size of a thread option).
With respect to claim 11, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein:
the first natural language model comprises a deep neural network and uses a probabilistic model to generate text (Amatriain-Rubio in [0069] and [0079] discloses first and second language model includes one or more neural network based machine learning model, first language model is constructed using a neural network based deep learning model architecture, embeddings are input to the neural network model, and the neural network model generates output based on the embeddings, first large language model is trained on a large dataset while second large language model is pre-trained on the same database but tune for a specific task type); or
the first natural language model and the second natural language model are substantially the same wherein the first natural language model and the second natural language model are substantially the same (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface; Amatriain-Rubio in [0069] and [0079] discloses first and second language model includes one or more neural network based machine learning model, first language model is constructed using a neural network based deep learning model architecture, embeddings are input to the neural network model, and the neural network model generates output based on the embeddings, first large language model is trained on a large dataset while second large language model is pre-trained on the same database but tune for a specific task type).
With respect to claim 13, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, further comprising executing the search query against a web-based set of documentation related to a software platform to obtain documentation data, wherein the second prompt is obtained using the second prompt generator based on the documentation data (Amatriain-Rubio in [0056] and [0057] discloses prompt based on documentation data).
With respect to claim 14, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein the result of the input query comprises a hyperlink, and wherein the hyperlink refers to additional information related to the result of the input query (Amatriain-Rubio in [0032] and [0080] discloses result including hyperlinks referring to additional information).
With respect to claim 15, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, further comprising:
displaying, on a screen of the client device, the result of the input query (Amatriain-Rubio in [0018] and [0047] discloses displaying results).
With respect to claim 16, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein:
generating the second text description data comprises generating, by a third natural language model, the second text description data (Amatriain-Rubio in [0025] and [0044] discloses a plurality of language models; McCurdy in [0024] and [0043] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired; McCurdy in [0041] and [0067] discloses generating responses using one or more language models; McCurdy in [0064] discloses queried vector database store vector representations of any relevant natural-language text information, forms, or templates useful for generating responses relevant to technical support, user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history including messages sent by the user and returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved);
generating the third text description data comprises generating, by the third natural language model, the third text description data (Amatriain-Rubio in [0025] and [0044] discloses a plurality of language models; McCurdy in [0024] and [0043] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired; McCurdy in [0041] and [0067] discloses generating responses using one or more language models; McCurdy in [0064] discloses queried vector database store vector representations of any relevant natural-language text information, forms, or templates useful for generating responses relevant to technical support, user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history including messages sent by the user and returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved); and
generating the fourth text description data comprises generating, by the third natural language model, the fourth text description data (Amatriain-Rubio in [0025] and [0044] discloses a plurality of language models; McCurdy in [0024] and [0043] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired; McCurdy in [0041] and [0067] discloses generating responses using one or more language models; McCurdy in [0064] discloses queried vector database store vector representations of any relevant natural-language text information, forms, or templates useful for generating responses relevant to technical support, user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history including messages sent by the user and returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved).
With respect to claim 17, Amatriain-Rubio in view of McCurdy discloses the method of claim 16, wherein the third natural language model is configured to generate summaries of input text data (Amatriain-Rubio in [0147] discloses generate summaries).
With respect to claim 18, Amatriain-Rubio in view of McCurdy discloses the method of claim 16, wherein constructing the plurality of vectorized databases further comprises:
converting the first text description data into first vectorized portions of text (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
storing the first vectorized portions of text in a first vectorized database of the plurality of vectorized databases (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
converting the second text description data into second vectorized portions of text (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
storing the second vectorized portions of text in a second vectorized database of the plurality of vectorized databases (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
converting the third text description data into third vectorized portions of text (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
storing the third vectorized portions of text in a third vectorized database of the plurality of vectorized databases (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model);
converting the fourth text description data into fourth vectorized portions of text (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model); and
storing the fourth vectorized portions of text in a fourth vectorized database of the plurality of vectorized databases (Amatriain-Rubio in [0025] and [0044] discloses language models; McCurdy in [0024] discloses databases storing descriptive user information, user device hardware information, user device software information, user account status information, technical support service level information, and/or any other information useful for identifying relevant user information and/or information describing a user device for which technical support may be desired. In some examples, databases can be databases of used by other servers and/or systems and can store additional data relevant to the task(s) performed by those servers and/or systems; McCurdy in [0064] discloses vector database queried can store vector representations of any relevant natural-language text information, the vector database can store forms or templates useful for generating responses relevant to technical support (e.g., a template outlining troubleshooting steps, etc.), user-performable instructions for resolving a technical support issue, and/or any other suitable pre-generated text that can be used by a machine-learning language model to generate responses useful for providing technical support and resolving technical support issues, the vector database can store vector representations of technical support chat history, chat history can include, for example, one or more messages sent by the user (e.g., as prompts) as well as messages returned by a human technical support technician as responses to the technical queries contained within those messages, chat histories stored by the vector database and retrieved can be specific to the user who submitted the prompt and/or other users who were provided with technical support by one or more human support technicians; McCurdy in [0067] discloses using language model).
With respect to claim 19, Amatriain-Rubio discloses a computing system (Amatriain-Rubio in [0255] and in Figure 8 discloses a computing system) comprising:
one or more processors (Amatriain-Rubio in [0255] and in Figure 8 discloses a computing system with one or more processors);
memory (Amatriain-Rubio in [0255] and in Figure 8 discloses a computing system with memory); and
program instructions, stored in the memory, that, upon execution by the one or more processors, cause the one or more processors to perform operations (Amatriain-Rubio in [0255] and in Figure 8 discloses a computing system with program instructions, stored in the memory executed by one or more processors to perform operations) comprising:
constructing a plurality of…databases… (Amatriain-Rubio in [0066] and [0202] discloses searching for and retrieving information by executing queries on databases, providing retrieved data from databases; Amatriain-Rubio in [0209] and [0210] discloses data stores implemented using databases, data written to and read from data stores, databases organize and store data records; here Amatriain-Rubio does not explicitly discloses a plurality of vectorized databases, but the McCurdy reference discloses the feature, as discussed below);
receiving, from a client device, an input query (Amatriain-Rubio in [0037] and [0054] discloses receiving a user input to a user assistance system from a user device);
obtaining, using a first prompt generator, a first prompt based on the input query (Amatriain-Rubio in [0044] and [0054] discloses user assistance system includes prompt generators and large language models, prompt generator formulates and outputs specific type of prompts that can be used as inputs to one or more large language models, prompt includes instructions to cause a large language model to generate an output in response to the prompt; Amatriain-Rubio in [0054] discloses user;
generating, by way of a first natural language model and based on the first prompt, a search query comprising a plurality of…values relating to the first prompt that are usable to retrieve…data from a plurality of…databases… (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio discloses generating a query using a natural language model and a prompt comprising a plurality of values to search a database, but does not explicitly disclose generating a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases, wherein the plurality of vectorized databases comprises vectorized portions of text relating to (i) summaries of features relating to a plurality of applications and (ii) descriptions of the plurality of applications, but the McCurdy reference discloses the feature, as discussed below);
executing, against the plurality of…databases, the search query using the plurality of…values to obtain the…data (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data, but the McCurdy reference discloses the feature, as discussed below);
obtaining, using a second prompt generator, a second prompt…(Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose prompt based on the vectorized data and the input query, but the McCurdy reference discloses the feature, as discussed below);
generating, by way of a second natural language model and based on the second prompt, a result of the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly discloses generating a result of a prompt based on vectorized data and the input query, but the McCurdy reference discloses the feature; and
transmitting the result of the input query to the client device (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data).
Amatriain-Rubio discloses constructing a plurality of databases, receiving an input query, obtaining a prompt from a prompt generator based on the input query, generating a search query comprising a plurality of values usable to retrieve data from a database using a natural language model and the prompt, executing the query against the database to obtain data, obtaining a second prompt using a second prompt generator, generating a result of an input query and transmitting the result to a client device, however Amatriain-Rubio does not explicitly disclose:
…a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications;
generating…a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from the plurality of vectorized databases;
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data;
obtaining…a second prompt based on the vectorized data and the input query;
generating…based on the second prompt, a result of the input query
transmitting the result of the input query….
The McCurdy reference discloses a plurality of vectorized databases, wherein the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications (McCurdy in [0005] and [0006] discloses a vector database comprising a plurality of vectors, each vector of the plurality of vectors representative of a text segment of a plurality of text segments, database configured to store vector embeddings representative of text segments; McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data; McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database; McCurdy in [0027]-[0029] discloses vector databases store vector information representative of natural-language text, vectors stored in vector database represent any suitable length of text, such as sentences, paragraphs, etc., storing vector embeddings useful for generating relevant responses, databases storing application data, the vectors representing sentences within messages and/entire messages sent through chat, and vector database can represent technical support chat histories, represent both user queries and responses, collection of chat history of messages including prompts and responses, vector database can separate chat history into messages and/or sentences and store vector embeddings of those text segments in vector database, which is summaries of features of chats and response related to technical support discussions; McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions);
generating a search query comprising a plurality of vector values relating to a first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
obtaining a second prompt based on the vectorized data and the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
generating based on the second prompt, a result of the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information); and
transmitting the result of the input query (McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, having the teachings of Amatriain-Rubio and McCurdy, to have combined Amatriain-Rubio and McCurdy. The motivation to combine Amatriain-Rubio and McCurdy would be to improve the accuracy and/or relevance of natural-language generated by a model in response to a prompt by augmenting prompts with additional information to provide context to the language model (McCurdy: [0004]).
With respect to claim 20, Amatriain-Rubio discloses a non-transitory machine-readable medium storing program instructions that, when executed by one or more processors of a computing system, cause the one or more processors to perform operations (Amatriain-Rubio in [0255] and in Figure 8 discloses a computing system with program instructions, stored in the memory executed by one or more processors to perform operations) comprising:
constructing a plurality of…databases… (Amatriain-Rubio in [0066] and [0202] discloses searching for and retrieving information by executing queries on databases, providing retrieved data from databases; Amatriain-Rubio in [0209] and [0210] discloses data stores implemented using databases, data written to and read from data stores, databases organize and store data records; here Amatriain-Rubio does not explicitly discloses a plurality of vectorized databases, but the McCurdy reference discloses the feature, as discussed below);
receiving, from a client device, an input query (Amatriain-Rubio in [0037] and [0054] discloses receiving a user input to a user assistance system from a user device);
obtaining, using a first prompt generator, a first prompt based on the input query (Amatriain-Rubio in [0044] and [0054] discloses user assistance system includes prompt generators and large language models, prompt generator formulates and outputs specific type of prompts that can be used as inputs to one or more large language models, prompt includes instructions to cause a large language model to generate an output in response to the prompt; Amatriain-Rubio in [0054] discloses user;
generating, by way of a first natural language model and based on the first prompt, a search query comprising a plurality of…values relating to the first prompt that are usable to retrieve…data from a plurality of…databases… (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio discloses generating a query using a natural language model and a prompt comprising a plurality of values to search a database, but does not explicitly disclose generating a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases, wherein the plurality of vectorized databases comprises vectorized portions of text relating to (i) summaries of features relating to a plurality of applications and (ii) descriptions of the plurality of applications, but the McCurdy reference discloses the feature, as discussed below);
executing, against the plurality of…databases, the search query using the plurality of…values to obtain the…data (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data, but the McCurdy reference discloses the feature, as discussed below);
obtaining, using a second prompt generator, a second prompt…(Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly disclose prompt based on the vectorized data and the input query, but the McCurdy reference discloses the feature, as discussed below);
generating, by way of a second natural language model and based on the second prompt, a result of the input query (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data; here Amatriain-Rubio does not explicitly discloses generating a result of a prompt based on vectorized data and the input query, but the McCurdy reference discloses the feature; and
transmitting the result of the input query to the client device (Amatriain-Rubio in [0024] discloses a large language model configured to perform natural language processing tasks; Amatriain-Rubio in [0066] and [0202] discloses retrieving data from database by executing queries; Amatriain-Rubio in [0203] discloses operating online dialogs with users, using one or more large language models, based on input received via thread-based user assistance interface, generating prompts based on input data).
Amatriain-Rubio discloses constructing a plurality of databases, receiving an input query, obtaining a prompt from a prompt generator based on the input query, generating a search query comprising a plurality of values usable to retrieve data from a database using a natural language model and the prompt, executing the query against the database to obtain data, obtaining a second prompt using a second prompt generator, generating a result of an input query and transmitting the result to a client device, however Amatriain-Rubio does not explicitly disclose:
…a plurality of vectorized databases, wherein constructing the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications;
generating…a search query comprising a plurality of vector values relating to the first prompt that are usable to retrieve vectorized data from the plurality of vectorized databases;
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data;
obtaining…a second prompt based on the vectorized data and the input query;
generating…based on the second prompt, a result of the input query
transmitting the result of the input query….
The McCurdy reference discloses a plurality of vectorized databases, wherein the plurality of vectorized databases comprises receiving first text description data comprising summaries of features relating to a plurality of applications, generating second text description data comprising descriptions of the plurality of applications, generating third text description data comprising descriptions of groupings of the plurality of applications, and generating fourth text description data comprising descriptions of sets of the groupings of the plurality of applications (McCurdy in [0005] and [0006] discloses a vector database comprising a plurality of vectors, each vector of the plurality of vectors representative of a text segment of a plurality of text segments, database configured to store vector embeddings representative of text segments; McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data; McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database; McCurdy in [0027]-[0029] discloses vector databases store vector information representative of natural-language text, vectors stored in vector database represent any suitable length of text, such as sentences, paragraphs, etc., storing vector embeddings useful for generating relevant responses, databases storing application data, the vectors representing sentences within messages and/entire messages sent through chat, and vector database can represent technical support chat histories, represent both user queries and responses, collection of chat history of messages including prompts and responses, vector database can separate chat history into messages and/or sentences and store vector embeddings of those text segments in vector database, which is summaries of features of chats and response related to technical support discussions; McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions);
generating a search query comprising a plurality of vector values relating to a first prompt that are usable to retrieve vectorized data from a plurality of vectorized databases (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
executing, against the plurality of vectorized databases, the search query using the plurality of vector values to obtain the vectorized data (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0014] and [0030] discloses queries made to one or more vector databases, and a server can include multiple vector databases storing and organizing vectors representative of different types of data, such as chat history data, template data, etc.; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
obtaining a second prompt based on the vectorized data and the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information);
generating based on the second prompt, a result of the input query (McCurdy in [0005] and [0031] discloses receiving a first natural language prompt from a user and a user identifier, the first prompt including at least one technical support query, querying a first database to retrieve first information, generating a first vector embedding representative of the first information and the first prompt, querying a second database using the first vector embedding to retrieve second information, generating a response text based on the first prompt, first information, and the second information, second database is a vector database comprising a plurality of vectors, plurality of vectors representative of a text segment of a plurality of text segments, the second information comprises at least one text segment, and the first natural language response text responsive to the at least one technical support query; McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user; McCurdy in [0085] and [0086] discloses querying software information and retrieving vector embedding representative of the information); and
transmitting the result of the input query (McCurdy in [0039] and [0043] discloses performing queries of vector databases and retrieving text string information from the vector database, retrieving user-specific information from vector database related to software information discussed during technical support session messages; McCurdy in [0055] discloses receiving a prompt and user identifier, querying a database with the user identifier, creating a query vector, querying a vector database, augmenting a natural language query with retrieved database data, generate natural language response with a language model, transmitting the response to a user device, displaying the response to the user).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, having the teachings of Amatriain-Rubio and McCurdy, to have combined Amatriain-Rubio and McCurdy. The motivation to combine Amatriain-Rubio and McCurdy would be to improve the accuracy and/or relevance of natural-language generated by a model in response to a prompt by augmenting prompts with additional information to provide context to the language model (McCurdy: [0004]).
With respect to claim 21, Amatriain-Rubio in view of McCurdy discloses the method of claim 1, wherein executing the search query comprises:
determining at least one of the plurality of vectorized databases is irrelevant to the search query (McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data; McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database; McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions; McCurdy in [0046] and [0051] discloses vectors of vector database include embedding dimensions that correspond to the types of retrievable data, using similarity in vector embedding dimensions of query vector and database vectors to narrow search space, accurately identifying relevant information, reducing the likelihood that irrelevant information is retrieved); and
upon determining the at least one database is irrelevant to the search query, skipping searching the at least one database when obtaining the vectorized data (McCurdy in [0014] and [0030] discloses using queries to retrieve information from multiple databases, queries made to one or more vector databases, multiple vector databases store and organize vectors representative of different types of data, each vector database stores and organizes vectors representative of a single type of data; McCurdy in [0024] and [0043] discloses databases storing descriptive information, such as user device software and hardware information, and any other information useful for identifying relevant information, specific information, such as software information used for identifying data stored to vector database; McCurdy in [0034] and [0049] discloses databases operate as vector database and can be partitions of a database, vector database partitioned into different partitions, using identifiers to identify one or more relevant partitions and query those partitions; McCurdy in [0046] and [0051] discloses vectors of vector database include embedding dimensions that correspond to the types of retrievable data, using similarity in vector embedding dimensions of query vector and database vectors to narrow search space, accurately identifying relevant information, reducing the likelihood that irrelevant information is retrieved).
Remarks
The relevant prior art of record that are not used in claim rejections but are pertinent to the claims or disclosure are:
Hensley (US Pub 2025/0251932), which discloses constructing a vector database that houses summaries.
Hasan (US Pat 9,965,793), which discloses searching for items in a first database instead of a second database based on a determination.
Turner (US Pub 2026/0113310), which discloses generating a vector representation databases by generating summaries.
Xiong (US Pat 11,256,707), which discloses searching for data based on a query in one or a plurality of database partitions based on relevancy.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to REZWANUL MAHMOOD whose telephone number is (571)272-5625. The examiner can normally be reached M-F 9-5:30.
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, Ann J. Lo can be reached at 571-272-9767. 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.
/R.M/Examiner, Art Unit 2159 /ANN J LO/Supervisory Patent Examiner, Art Unit 2159