DETAILED ACTION
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 .
This Office Action is in response to correspondence filed 11 February 2025 in reference to application 19/050,613. Claims 1-20 are pending and have been examined.
Claim Objections
Claims 3 and 16 objected to because of the following informalities: “participant” should be “participate”. Appropriate correction is required.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 14, and 20 recite receiving a prompt for the LLM; determining a classification based on metadata that is associated with the prompt; retrieving a set of trackers that is defined for the classification; generating a context-specific output by modifying execution of the prompt by the LLM based on context that the set of trackers add to the prompt, wherein modifying the execution comprises selecting between different paths that lead to a plurality of different outputs for the prompt in the LLM based on the context that the set of trackers add to the prompt; and performing an automated action in response to the prompt based on the context-specific output.
The limitation of receiving a prompt for the LLM, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “computer implemented” in claim 1, “one or more hardware processors” is claim 14, and “a non-transitory computer-readable medium” in claim 20, nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “by a processor” language, “receiving” in the context of this claim encompasses a person reading a prompt designated for an LLM.
The limitation of determining a classification based on metadata that is associated with the prompt, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the computer components, “determining” in the context of this claim encompasses a person reading metadata and thinking of a classification.
The limitation of retrieving a set of trackers that is defined for the classification, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the computer components, “retrieving” in the context of this claim encompasses a person selecting trackers based on the classification.
The limitation of generating a context-specific output by modifying execution of the prompt by the LLM based on context that the set of trackers add to the prompt, wherein modifying the execution comprises selecting between different paths that lead to a plurality of different outputs for the prompt in the LLM based on the context that the set of trackers add to the prompt, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the computer components, “generating” in the context of this claim encompasses a person modifying the prompts based on context read from the trackers in order to select different paths of output expected from the LLM.
The limitation of performing an automated action in response to the prompt based on the context-specific output is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, but for the computer components, “performing” in the context of this claim encompasses a person providing a response to a user based on the output read from the interface of an LLM.
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.
This judicial exception is not integrated into a practical application. In particular, the claims only additionally recite “computer implemented” in claim 1, “one or more hardware processors” is claim 14, and “a non-transitory computer-readable medium” in claim 20. The these components are recited at a high-level of generality such that it amounts no more than mere instructions to apply the exception using a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of computer components amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claims are not patent eligible.
Claims 2 and 15 further recite defining the metadata based on a role of a user that submits the prompt. However a person can perform this step by reading the roll of the user and define the metadata based on that roll. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore these claims are not patent eligible.
Claims 3 and 16 further recite defining the metadata based on one or more users that participant in a communication identified in the prompt. However a person can perform this step by reading the names of participants in a communication and define the metadata based on the users. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore these claims are not patent eligible.
Claims 4 and 17 further recite presenting an interface comprising different fields for adding, removing, and modifying the set of trackers associated with the classification. However a person can perform this step creating a form for adding, removing, and modifying trackers. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore these claims are not patent eligible.
Claims 5 and 18 further recite defining a plurality of classifications in a graphical user interface; associating the plurality of classifications to different user roles or types of content; and defining a different set of trackers for each classification of the plurality of classifications based on inputs provided in the graphical user interface. However a person can perform this step creating a form for selecting classifications, thinking of what rolls are associated with classifications, writing trackers based on each classification on inputs based on received inputs in the form. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore these claims are not patent eligible.
Claims 6 and 19 further recite generating a modified prompt by appending the set of trackers to the prompt prior to executing the prompt with the LLM; and issuing the modified prompt instead of the prompt as input to the LLM. However a person can perform this step by writing a new prompt by adding the tracker and typing the prompt into the interface of an LLM. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore these claims are not patent eligible.
Claim 7 further recites dynamically enhancing the prompt with first context defined by a first set of trackers in response to determining a first classification; and dynamically enhancing the prompt with second context defined by a second set of trackers in response to determining a second classification. However a person can perform this step adding context defined by a first and second tracker from two classifications. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 8 further recites generating an answer to a question in the prompt with data from a particular domain associated with the classification and by excluding data from other domains that are associated with other classifications. However a person can perform this step by removing domains by specifying that in the prompt. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 9 further recites selecting a response from a plurality of responses that replies to a first contextual interpretation of the prompt based on the classification, and wherein each other response from the plurality of response is a reply to a different contextual interpretation of the prompt. However a person can perform this step by trying different context in prompts and selecting the appropriate generated output from the interface of an LLM. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 10 further recites determining one or more features of the prompt that ambiguously reference different data in different domains; and resolving the one or more features to a particular domain from the different domains based on the context that the set of trackers add to the prompt.. However a person can perform this step by observing ambiguity in the prompt and replacing the ambiguity with context read from the tracker. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 11 further recites activating a chatbot; and responding directly to the prompt with the context-specific output from the chatbot. However a person can perform this step by opening the interface of a chatbot and reading the output of the chatbot. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 12 further recites extracting a first set of features from audio content of a conference identified in the prompt based a relevance between the first set of features and a first set of trackers defined for a first classification; and extracting a different second set of features from the audio content based a relevance between the different second set of features and a second set of trackers defined for a second classification.. However a person can perform this step by of listening to audio and extracting content and comparing the context to the trackers. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim 13 further recites training the LLM with data from different departments of an organization; and determining that the prompt relates to a particular department of the different departments based on the context that the set of trackers add to the prompt; and generating a response to the prompt using the data from the particular department in response to determining that the prompt relates to the particular department.. However a person can perform this step by of providing training data to the interface of an LLM, reading a prompt to determine a particular department, and writing a response according to the department. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. Therefore the claim is not patent eligible.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-3, 5-9, 11, 14-16, and 18-20 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Sharma et al. (US PAP 2026/0064870).
Consider claim 1, Sharma teaches A computer-implemented method for dynamically adapting a Large Language Model (LLM) for targeted output in different domains (abstract), the computer-implemented method comprising:
receiving a prompt for the LLM (0034, 0063, prompt input by user);
determining a classification based on metadata that is associated with the prompt (0035, classification of user roll from user profile, i.e. metadata);
retrieving a set of trackers that is defined for the classification (0037-38, 0066, determining text blocks, which at 0028-29 can be documents of the organization );
generating a context-specific output by modifying execution of the prompt by the LLM based on context that the set of trackers add to the prompt, wherein modifying the execution comprises selecting between different paths that lead to a plurality of different outputs for the prompt in the LLM based on the context that the set of trackers add to the prompt (0067-68 generating modified prompt including only context relevant to the user position, thus generating output relevant to the user position. Also see figure 8 for modified prompt); and
performing an automated action in response to the prompt based on the context-specific output (0067, generating a response to the user).
Consider claim 2, Sharma teaches the computer-implemented method of claim 1, further comprising: defining the metadata based on a role of a user that submits the prompt (0035, classification of user roll from user profile, i.e. metadata).
Consider claim 3, Sharma teaches the computer-implemented method of claim 1, further comprising: defining the metadata based on one or more users that participant in a communication identified in the prompt (0028, memos, letters, emails and the people who wrote them).
Consider claim 5, Sharma teaches The computer-implemented method of claim 1, further comprising:
defining a plurality of classifications in a graphical user interface (0035, user rolls, may be displayed in UI);
associating the plurality of classifications to different user roles or types of content (0055-0059, determining text blocks of data that are associated with positions); and
defining a different set of trackers for each classification of the plurality of classifications based on inputs provided in the graphical user interface (0055-0059, determining text blocks of data that are associated with positions and selecting them for inclusion in modified prompts).
Consider claim 6, Sharma teaches the computer-implemented method of claim 1, further comprising:
generating a modified prompt by appending the set of trackers to the prompt prior to executing the prompt with the LLM (0067, generating a modified prompt); and
issuing the modified prompt instead of the prompt as input to the LLM (0067-68 modified prompt sent to LLM).
Consider claim 7, Sharma teaches The computer-implemented method of claim 1, further comprising:
dynamically enhancing the prompt with first context defined by a first set of trackers in response to determining a first classification (0067-68 generating modified prompt including only context relevant to the user position, geologist for example); and
dynamically enhancing the prompt with second context defined by a second set of trackers in response to determining a second classification (0067-68 generating modified prompt including only context relevant to the user position, project manager for example ).
Consider claim 8, Sharma teaches the computer-implemented method of claim 1, wherein generating the context-specific output comprises: generating an answer to a question in the prompt with data from a particular domain associated with the classification and by excluding data from other domains that are associated with other classifications (0058, 0068, data can be excluded from prompt if the topic is not relevant enough to position description ).
Consider claim 9, Sharma teaches The computer-implemented method of claim 1, wherein generating the context-specific output comprises: selecting a response from a plurality of responses that replies to a first contextual interpretation of the prompt based on the classification, and wherein each other response from the plurality of response is a reply to a different contextual interpretation of the prompt (0067-68 generating modified prompt including only context relevant to the user position, thus generating output relevant to the user position. Also see figure 8 for modified prompt based on different contexts).
Consider claim 11, Sharma teaches The computer-implemented method of claim 1, wherein performing the automated action comprises: activating a chatbot; and responding directly to the prompt with the context-specific output from the chatbot (0025, user may enter queries and receive responses in chat interface).
Consider claim 14, Sharma teaches a system for dynamically adapting a Large Language Model (LLM) for targeted output in different domains (abstract), the system comprising:
one or more hardware processors (0083, processors) configured to:
receive a prompt for the LLM (0034, 0063, prompt input by user);
determine a classification based on metadata that is associated with the prompt (0035, classification of user roll from user profile, i.e. metadata);
retrieve a set of trackers that is defined for the classification (0037-38, 0066, determining text blocks, which at 0028-29 can be documents of the organization );
generate a context-specific output by modifying execution of the prompt by the LLM based on context that the set of trackers add to the prompt, wherein modifying the execution comprises selecting between different paths that lead to a plurality of different outputs for the prompt in the LLM based on the context that the set of trackers add to the prompt (0067-68 generating modified prompt including only context relevant to the user position, thus generating output relevant to the user position. Also see figure 8 for modified prompt); and
perform an automated action in response to the prompt based on the context-specific output (0067, generating a response to the user).
Claim 15 contains similar limitations as claim 2 and therefore is rejected for the same reasons.
Claim 16 contains similar limitations as claim 3 and therefore is rejected for the same reasons.
Claim 18 contains similar limitations as claim 5 and therefore is rejected for the same reasons.
Claim 19 contains similar limitations as claim 6 and therefore is rejected for the same reasons.
Consider claim 20, A non-transitory computer-readable medium storing program instructions that, when executed by one or more hardware processors of a system that dynamically adapts a Large Language Model (LLM) for targeted output in different domains (abstract, 0082,83, computer readable media and processor), cause the system to perform operations comprising:
receiving a prompt for the LLM (0034, 0063, prompt input by user);
determining a classification based on metadata that is associated with the prompt (0035, classification of user roll from user profile, i.e. metadata);
retrieving a set of trackers that is defined for the classification (0037-38, 0066, determining text blocks, which at 0028-29 can be documents of the organization );
generating a context-specific output by modifying execution of the prompt by the LLM based on context that the set of trackers add to the prompt, wherein modifying the execution comprises selecting between different paths that lead to a plurality of different outputs for the prompt in the LLM based on the context that the set of trackers add to the prompt (0067-68 generating modified prompt including only context relevant to the user position, thus generating output relevant to the user position. Also see figure 8 for modified prompt); and
performing an automated action in response to the prompt based on the context-specific output (0067, generating a response to the user).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 4, 10, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sharma in view of Zhang et al. (US PAP 2026/0086832).
Consider claim 4, Sharma teaches The computer-implemented method of claim 1, but does not specifically teach presenting an interface comprising different fields for adding, removing, and modifying the set of trackers associated with the classification.
In the same field of LLM based support, Zhang teaches presenting an interface comprising different fields for adding, removing, and modifying the set of trackers associated with the classification (figures 3A-D, 0065, 0079-84, user can add and remove information, documents, and files etc associated with user profile).
It would have been obvious to one of ordinary skill in the art at the time of effective filing to provide an interface for users to provide context as taught by Zhang in the system of Sharma in order to allow the user to provide information that is relevant to answer user queries.
Consider claim 10, Sharma teaches the computer-implemented method of claim 1, but does not specifically teach wherein generating the context-specific output comprises:
determining one or more features of the prompt that ambiguously reference different data in different domains; and
resolving the one or more features to a particular domain from the different domains based on the context that the set of trackers add to the prompt.
In the same field of LLM based support, Zhang teaches
determining one or more features of the prompt that ambiguously reference different data in different domains (0169, demonstrative pronouns etc); and
resolving the one or more features to a particular domain from the different domains based on the context that the set of trackers add to the prompt (0169, replacing with unambiguous identification).
It would have been obvious to one of ordinary skill in the art at the time of effective filing to disambiguate prompts as taught by Zhang in the system of Sharma in order to better fulfill user intent in the prompt response.
Claim 17 contains similar limitations as claim 4 and therefore is rejected for the same reasons.
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sharma in view of Inkpen et al. (US PAP 2025/0157103).
Consider claim 12, Sharma teaches the computer-implemented method of claim 1, but does not specifically teach wherein generating the context-specific output comprises:
extracting a first set of features from audio content of a conference identified in the prompt based a relevance between the first set of features and a first set of trackers defined for a first classification; and
extracting a different second set of features from the audio content based a relevance between the different second set of features and a second set of trackers defined for a second classification.
In the same field of LLM assistance, Inkpen teaches
extracting a first set of features from audio content of a conference identified in the prompt based a relevance between the first set of features and a first set of trackers defined for a first classification (0038, 0048, 0051, extracting transcriptions for particular rolls, such as CEO etc, ); and
extracting a different second set of features from the audio content based a relevance between the different second set of features and a second set of trackers defined for a second classification (0038, 0048, 0051, extracting transcriptions for particular rolls, such as CEO and other users).
It would have been obvious to one of ordinary skill in the art at the time of effective filing to process conference data as taught by Inkpen as part of the knowledge base in Sharma in order to expand the knowledge base that is available for query processing.
Allowable Subject Matter
Claim 13 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 101, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims. The following is a statement of reasons for the indication of allowable subject matter:
Consider claim 13, Sharma teaches The computer-implemented method of claim 1. However the prior art of record does not specifically teach “training the LLM with data from different departments of an organization; and wherein generating the context-specific output comprises: determining that the prompt relates to a particular department of the different departments based on the context that the set of trackers add to the prompt; and generating a response to the prompt using the data from the particular department in response to determining that the prompt relates to the particular department.” when combined with each and every other limitation of the claim. Rather, in Sharma, users are limited to data according to their job description and not able to access data by specifying other departments in their prompt as claimed. Therefore claim 13 contains allowable subject matter.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Wilson et al. (US PAP 2026/003481) teaches a similar method of LLM classification methods.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DOUGLAS C GODBOLD whose telephone number is (571)270-1451. The examiner can normally be reached 6:30am-5pm Monday-Thursday.
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, Andrew Flanders can be reached at (571)272-7516. 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.
DOUGLAS GODBOLD
Examiner
Art Unit 2655
/DOUGLAS GODBOLD/Primary Examiner, Art Unit 2655
Prosecution Insights