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 12 November 2024 in reference to application 18/944,852. Claims 1-20 are pending and have been examined.
Claim Objections
Claim 4 is objected to because of the following informalities: In line 3 of the claim, “mode” should be “model”. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 9 and 10 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 9 recites the limitation "the human to computer dialog" in line 1. There is insufficient antecedent basis for this limitation in the claim. Therefore claim 9 is indefinite
Claim 10 depends on claim 9 and therefore is indefinite as well.
For purposes of examination, it will be assumed these claims depend on claim 8.
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 19-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1 and 19 recite receiving a user query; in response to receiving the user query, generating a tool-use representation that includes one or more reasoning blocks, based on processing one or more prompts, respectively, using a generative model, wherein the one or more prompts include a first prompt determined from the user query and metadata associated with a list of application programming interfaces (APIs), wherein the one or more reasoning blocks includes a tool call reasoning block that identifies: an identifier of an API, one or more parameters of the API, and one or more parameter values for the one or more parameters of the API; generating, based on processing the tool-use representation that includes the one or more reasoning blocks, a response to the user query; and causing the response to be rendered in response to the user query.
The limitation of receiving a user query, 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 “one or more processors,” and “a memory storing instructions,” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the computer components, “receiving” in the context of this claim encompasses a person listening to a user speak a query.
The limitation of generating a tool-use representation that includes one or more reasoning blocks, based on processing one or more prompts, respectively, wherein the one or more prompts include a first prompt determined from the user query and metadata associated with a list of application programming interfaces (APIs), wherein the one or more reasoning blocks includes a tool call reasoning block that identifies: an identifier of an API, one or more parameters of the API, and one or more parameter values for the one or more parameters of the API, 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 the person reading a prompt and data about available APIs, and writing out API calls and associated parameters and instructions to use them. This limitation additionally recites “using a generative model” also amount to a generic computer component as no generative models are well known and no specific details of the model are provided.
The limitation of generating, based on processing the tool-use representation that includes the one or more reasoning blocks, a response to the user query, 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 the person reading results generated from an APL along with the instructions to use the API and writing a response to the user query.
The limitation of causing the response to be rendered in response to the user query, 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, “causing” in the context of this claim encompasses the writing out a response to the user query and handing it to the user.
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 processors, memory, and a generative model. 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, these additional elements do 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 processors, memories, and generative models 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.
Claim 2 further recites the one or more reasoning blocks further include a text reasoning block that includes the one or more parameter values. However a person could write out text blocks that include parameter values. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 3 further recites the text reasoning block is generated prior to generating the tool-use representation. However a person could write out text blocks that include parameter values before writing out API calls. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 4 further recites processing the first prompt to generate a first model output from which the text reasoning block is derived; and processing a second prompt to generate a second model output from which the tool call reasoning block is derived, wherein the second prompt is determined from the first prompt and the text reasoning block. However a person could determine the text block first, create a second prompt, and create a tool call based on the second prompt and the text block. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 5 further recites wherein the tool call reasoning block further includes an execution result acquired based on execution of the API using the one or more parameter values. However a person could read the result of an API call and write out a result. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 6 further recites wherein the tool call reasoning block further includes an indication that indicates end of reasoning, and generating the response to the user query is performed in response to detecting the indication that indicates end of reasoning in the tool-use representation. However a person could write out an indication of the end of instructions, and generate a response based on reading the end of instructions indication. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 7 further recites wherein the user query includes a request to perform a mathematical operation, and wherein the API is associated with a python code executor. However a person could read a request to solve an equation and write an API call associated with a python code executor. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 8 further recites the user request is determined from a human-to-computer dialog having one or more turns of user input. However a person could read a read a transcript of a human to computer dialog. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 9 further recites the human-to-computer dialog further includes one or more turns of assistant input that are generated using an LLM-based assistant that accesses the generative model. However a person could read a read a transcript of a human to computer dialog. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 10 further recites the user query includes a request to access a tool external to the LLM-based assistant, wherein the tool is associated with one or more APIs. However a person could read a read a request to use a tool associated with an API. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 11 further recites determining whether the user query identifies any tool-use task to be performed using one or more APIs, wherein generating the one or more reasoning blocks is in response to determining that the user query identifies a tool-use task. However a person could read a read a request to use a tool associated with an API and write logic to call the API. Similar to above, no additional elements are recited that provide a practical application or amount to significantly more than the abstract idea. This claim is not patent eligible.
Claim 20 contains the same limitations as claims 2 and 3 and therefore is not patent eligible for the same reasons.
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) 1-6, 8-17, 19, and 20 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Shi et al. (US Patent 12,586,578).
Consider claim 1, Shi teaches A method implemented using one or more processors (abstract, col 55 line 52-57, CPUs), the method comprising:
receiving a user query (col 8 lines 20-25, user input data, line 60 – col 9 line 12, different tasks queries from user);
in response to receiving the user query, generating a tool-use representation that includes one or more reasoning blocks, based on processing one or more prompts, respectively, using a generative model (col 30 lines 49-col 32 line 4, using language model to generate API request based on prompts),
wherein the one or more prompts include a first prompt determined from the user query and metadata associated with a list of application programming interfaces (APIs) (see col 30 line 65- col 31 line 32, example prompts, which include a task that has been generated from user query, and a list of relevant APIs),
wherein the one or more reasoning blocks includes a tool call reasoning block that identifies: an identifier of an API, one or more parameters of the API, and one or more parameter values for the one or more parameters of the API (col 31 lines 42-65, generating API calls including parameter values, col 2 lines 60-67, example API call including parameters and identifier);
generating, based on processing the tool-use representation that includes the one or more reasoning blocks, a response to the user query (col 32 line 1- col 33 line 15, generating responses to the user query using API calls); and
causing the response to be rendered in response to the user query (col 38 lines 9-30, TTS or visual data generated for user).
Consider claim 2, Shi teaches the method of claim 1, wherein the one or more reasoning blocks further include a text reasoning block that includes the one or more parameter values (col 25 lines 40- col 27 line 40, generating task plan describing tasks that should be completed to fulfill user query, including parameters like which lights to turn on etc. ).
Consider claim 3, Shi teaches the method of claim 2, wherein the text reasoning block is generated, using the generative model, prior to generating the tool-use representation (Col 25 lines 40- col 27 line 40, Col 30 lines 49-col 32 line 4, task plan is used to generate API calls and thus must be generated first).
Consider claim 4, Shi teaches wherein generating the tool-use representation that includes the one or more reasoning blocks comprises:
processing the first prompt, using the generative mode, to generate a first model output from which the text reasoning block is derived (Col 25 lines 40- col 27 line 40, generating task plan describing tasks that should be completed to fulfill user query, using a prompt); and
processing a second prompt, using the generative model, to generate a second model output from which the tool call reasoning block is derived, wherein the second prompt is determined from the first prompt and the text reasoning block (col 30 line 65- col 31 line 32, example prompts, which include a task that has been generated from task plan, col 2 line 50, “one or more LLMs,” so each of the LLM tasks could be implemented using the same LLM).
Consider claim 5, Shi teaches the method of claim 1, wherein the tool call reasoning block further includes an execution result acquired based on execution of the API using the one or more parameter values (col 32 lines 4-20, causing API calls to be executed and returning results).
Consider claim 6, Shi teaches the method of claim 1, wherein the tool call reasoning block further includes an indication that indicates end of reasoning, and generating the response to the user query is performed in response to detecting the indication that indicates end of reasoning in the tool-use representation (col 29-3-37, task selection LM processes tasks and determine order in which they should be executed to fulfill user request. Thus the last task would be the end of reasoning and would result in the request being fulfilled.).
Consider claim 8, Shi teaches the method of claim 1, wherein the user request is determined from a human-to-computer dialog having one or more turns of user input (col 12 lines 5-40, dialog history, human inputs and computer responses).
Consider claim 9, Shi teaches the method of claim 2, wherein the human-to-computer dialog further includes one or more turns of assistant input that are generated using an LLM-based assistant that accesses the generative model (col 12 lines 5-40, dialog history, human inputs and computer responses, abstract, LLM based system).
Consider claim 10, Shi teaches the method of claim 9, wherein the user query includes a request to access a tool external to the LLM-based assistant, wherein the tool is associated with one or more APIs (col 8 lines 60-67, examples of user requests that request using APIs).
Consider claim 11, Shi teaches the method of claim 1, further comprising: determining whether the user query identifies any tool-use task to be performed using one or more APIs, wherein generating the one or more reasoning blocks is in response to determining that the user query identifies a tool-use task (col 29 line 45-67, generating a list of potentially relevant APIs before determining an API call).
Consider claim 12, Shi teaches A method implemented using one or more processors (abstract, col 55 line 52-57, CPUs), the method comprising:
receiving one or more user inputs, wherein the one or more user inputs are provided via a user interface of an LLM-based assistant accessible via a client device, and wherein the LLM-based assistant accesses a generative model and a list of application programming interfaces (APIs) (col 8 lines 20-25, user input data, line 60 – col 9 line 12, different tasks queries from user, col 2 lines 39-67, LLM system that can generate API calls, figure 9, various client devices 110a-n);
determining that the one or more user inputs indicate a request to perform an application action via an application that is external to the LLM-based assistant (col 29 line 45-67, generating a list of potentially relevant APIs to user query);
in response to receiving the user query and in response to determining that the one or more user inputs indicate the request to perform the application action, generating a tool-use representation that includes one or more reasoning blocks, based on processing one or more prompts, respectively, using the generative model (col 30 lines 49-col 32 line 4, using language model to generate API request based on prompts,),
wherein the one or more prompts include a first prompt determined from the user query and metadata associated with a list of application programming interfaces (APIs) (see col 30 line 65- col 31 line 32, example prompts, which include a task that has been generated from user query, and a list of relevant APIs),
wherein the one or more reasoning blocks includes a tool call reasoning block that identifies: an identifier of an API, one or more parameters of the API, and one or more parameter values for the one or more parameters of the API (col 31 lines 42-65, generating API calls including parameter values, col 2 lines 60-67, example API call including parameters and identifier);
generating, based on processing the tool-use representation that includes the one or more reasoning blocks, a response to the user query (col 32 line 1- col 33 line 15, generating responses to the user query using API calls); and
causing the response to be rendered in response to the user query (col 38 lines 9-30, TTS or visual data generated for user).
Claim 13 contains similar subject matter as claim 2 and is rejected for the same reasons.
Claim 14 contains similar subject matter as claim 3 and is rejected for the same reasons.
Claim 15 contains similar subject matter as claim 4 and is rejected for the same reasons.
Claim 16 contains similar subject matter as claim 5 and is rejected for the same reasons.
Claim 17 contains similar subject matter as claim 6 and is rejected for the same reasons.
Consider claim 20, Shi teaches A system (abstract) comprising one or more processors (col 55 line 52-57, CPUs) and memory (col 56 lines 1-10, RAM, ROM, etc) storing instructions that, when executed, causes the one or more processors to:
receive a user query (col 8 lines 20-25, user input data, line 60 – col 9 line 12, different tasks queries from user);
in response to receiving the user query, generate a tool-use representation that includes one or more reasoning blocks, based on processing one or more prompts, respectively, using a generative model (col 30 lines 49-col 32 line 4, using language model to generate API request based on prompts),
wherein the one or more prompts include a first prompt determined from the user query and metadata associated with a list of application programming interfaces (APIs) (see col 30 line 65- col 31 line 32, example prompts, which include a task that has been generated from user query, and a list of relevant APIs),
wherein the one or more reasoning blocks includes a tool call reasoning block that identifies: an identifier of an API, one or more parameters of the API, and one or more parameter values for the one or more parameters of the API (col 31 lines 42-65, generating API calls including parameter values, col 2 lines 60-67, example API call including parameters and identifier);
generate, based on processing the tool-use representation that includes the one or more reasoning blocks, a response to the user query (col 32 line 1- col 33 line 15, generating responses to the user query using API calls); and
cause the response to be rendered in response to the user query (col 38 lines 9-30, TTS or visual data generated for user).
Claim 20 contains similar subject matter as claims 2 and 3 and is rejected for the same reasons.
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) 7 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shi in view of Nori et al. (US PAP 2025/0148220) and further in view of Makhija et al. (US PAP 2026/0087447).
Consider claim 7, Shi teaches The method of claim 1, but does not specifically teach wherein the user query includes a request to perform a mathematical operation.
In the same field of LLM agents, Nori teaches the user query includes a request to perform a mathematical operation (0032, the prompt may be to solve a mathematical operation).
It would have been obvious to one of ordinary skill in the art at the time of effective filing to allow for requests to complete mathematical operations as taught by Nori in the system of Shi in order to allow the user to more easily solve math equations, increasing user convenience.
Shi and Nori do not specifically teach wherein the API is associated with a python code executor.
IN the same field of calling APIs, Makhija teaches wherein the API is associated with a python code executor (0070,0073, tools can be implemented using APIs and phyton functions).
It would have been obvious to one of ordinary skill in the art at the time of effective filing to user python as taught by Makhija in the system of Shi and Nori in order to use a well-known and accessible programing language.
Claim 18 contains similar subject matter as claim 7 and is rejected for the same reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Martini et al. (US PAP 2026/0087374) and Spoolin (US PAP 2026/0023931) teach similar methods of using LLMs to call API tools.
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