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 .
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.
Claim 6-10, and 16-20 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.
Claims 6 and 16 recite the limitation "the first neural network" and "the fourth network" in line 4 of the claim. There is insufficient antecedent basis for this limitation in the claim. Claim 1 recites "a first neural network model" and claim 6 recites "a fourth neural network model"; it is unclear whether the recited terms refer to these models or to different elements.
Claims 7, 8, 17, and 18 recite the limitation "the generating, jointly by the first neural network and the fourth neural network, the first sub-task package" in line 1 of the claim. There is insufficient antecedent basis for this limitation in the claim. Parent claims 6 and 16 recite generating by the first neural network or the fourth network (in the alternative), not jointly.
Further regarding claims 7 and 17, claim 7 recites "generating, by the fourth neural network model, the first sub-task from the initial first sub-task," while claim 1 recites "generating, by the first neural network model, the first sub-task from the task instruction." It is unclear how the first sub-task can be generated both by the first neural network model from the task instruction and by the fourth neural network model from the initial first sub-task. Claim 17 conflicts with claim 11 in the same manner.
Regarding claims 8 and 18, the claims recite "selecting, by the first neural network model or the second neural network model, the second neural network model from the plurality of the neural network models based on the first sub-task." The claim is indefinite because, in one alternative, the second neural network model selects itself before it has been selected, and it is further unclear whether this selecting step is the same as or in addition to the selecting step recited in claims 1 and 11.
Claims 9 and 19 recite the limitation "the sixth neural network model" in line 10 of the claim. There is insufficient antecedent basis for this limitation in the claim. The claims previously recite only "a sixth neural network."
Claim 10 recites the limitation "the first neural network" and "the second neural network" in line 1 of the claim. There is insufficient antecedent basis for this limitation in the claim. Claim 1 recites "a first neural network model" and "a second neural network model."
Claim 20 recites the limitation "the plurality of the neural network models" in line 6 of the claim. There is insufficient antecedent basis for this limitation in the claim. Claim 20 does not previously recite a plurality of neural network models.
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.
Claim 1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
MPEP 2106 (III) sets out steps for evaluating whether a claim is drawn to patent-eligible subject matter. The analysis of claims 1-20, in accordance with these steps, follows.
Step 1 Analysis:
Claims 1-10 are directed to method (processes). Claims 11-19 are directed to a system (machine). Claim 20 is directed to a computer program product (article of manufacture). Therefore, claims 1-20 fall into one of four statutory categories (i.e., process, machine, article of manufacture).
As to claim 1,
Step 2A Prong 1: this claim recites the following abstract ideas:
generating ... a first sub-task from the task instruction; (the limitation describes devising a sub-task by breaking the task instruction down into a smaller task, which is a mental process implemented using a pen and paper.)
selecting a second neural network model from the plurality of the neural network models based on the first sub-task; (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
generating ... a first sub-task package in a format compliant with the second neural network model; (the limitation describes writing out the sub-task in a required format, which is a mental process implemented using a pen and paper.)
generating ... a second sub-task based on the task instruction and the first output; (the limitation describes devising a further sub-task by evaluating the task instruction together with the received output, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
receiving, via a data interface, a task instruction; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by a first neural network model; by the first neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via the first connection, a first output from the second neural network model that executes the first sub-task package; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
causing the task instruction to be jointly performed by one or more selected neural network models from the plurality of neural network models based at least in part on the second sub-task. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 11,
Step 2A Prong 1: this claim recites the following abstract ideas:
generating ... a first sub-task from the task instruction; (the limitation describes devising a sub-task by breaking the task instruction down into a smaller task, which is a mental process implemented using a pen and paper.)
selecting a second neural network model from the plurality of the neural network models based on the first sub-task; (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
generating ... a first sub-task package in a format compliant with the second neural network model; (the limitation describes writing out the sub-task in a required format, which is a mental process implemented using a pen and paper.)
generating ... a second sub-task based on the task instruction and the first output; (the limitation describes devising a further sub-task by evaluating the task instruction together with the received output, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
a memory that stores the plurality of neural network models and a plurality of processor executable instructions; a communication interface that receives a task instruction; and one or more hardware processors that read and execute the plurality of processor-executable instructions from the memory to perform operations; (This limitation is directed to mere instruction to store the abstract idea on a generic memory and apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
a communication interface that receives a task instruction; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by a first neural network model; by the first neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via the first connection, a first output from the second neural network model that executes the first sub-task package; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
causing the task instruction to be jointly performed by one or more selected neural network models from the plurality of neural network models based at least in part on the second sub-task. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 20,
Step 2A Prong 1: this claim recites the following abstract ideas:
generating ... a first sub-task from the task instruction; (the limitation describes devising a sub-task by breaking the task instruction down into a smaller task, which is a mental process implemented using a pen and paper.)
selecting a second neural network model from the plurality of the neural network models based on the first sub-task; (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
generating ... a first sub-task package in a format compliant with the second neural network model; (the limitation describes writing out the sub-task in a required format, which is a mental process implemented using a pen and paper.)
generating ... a second sub-task based on the task instruction and the first output; (the limitation describes devising a further sub-task by evaluating the task instruction together with the received output, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
A non-transitory machine-readable medium comprising a plurality of machine-executable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform operations; (This limitation is directed to mere instruction to store the abstract idea on a generic memory and apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via a data interface, a task instruction; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by a first neural network model; by the first neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via the first connection, a first output from the second neural network model that executes the first sub-task package; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
causing the task instruction to be jointly performed by one or more selected neural network models from the plurality of neural network models based at least in part on the second sub-task. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 2 and 12,
Step 2A Prong 1: those claims recite the following abstract ideas:
selecting a third neural network model from the plurality of neural network models based on the second sub-task; (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
generating ... a second sub-task package in a format compliant with the third neural network model; (the limitation describes writing out the sub-task in a required format, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
building a second connection, via a second API between the first neural network model and the third neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by the first neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via the second connection, a second output from the third neural network model that executes the second sub-task package. (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 3 and 13,
Step 2A Prong 1: those claims recite the following abstract ideas:
wherein the first sub-task package comprises a first prompt compliant with the second neural network model, instructing the second neural network model to perform the first sub-task. (the limitation describes the content of the sub-task package being generated, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claims are ineligible.
As to claims 4 and 14,
Step 2A Prong 1: those claims recite the following abstract ideas:
wherein the first output comprises one or more of a completion status corresponding to the first sub-task, identification information for the first neural network model, or identification information for the second neural network model. (the limitation describes the content of the information being considered, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claims are ineligible.
As to claims 5 and 15,
Step 2A Prong 1: those claims recite the following abstract ideas:
generating ... the first sub-task based on the task instruction and the human instruction. (the limitation describes devising a sub-task by evaluating the task instruction together with a received instruction, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
The additional limitation of claim 15 "wherein the operations further comprise" (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, by one of the first neural network model or the second neural network model, a human instruction; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by the first neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 6 and 16,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claims depend on claims 1 and 11.
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
The additional limitation of claim 16 "wherein the operations further comprise" (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
a fourth neural network model communicatively coupled to the first neural network model via a third connection based on a third API; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
generating, by the first neural network or the fourth network, the first sub-task package. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 7 and 17,
Step 2A Prong 1: those claims recite the following abstract ideas:
generating ... an initial first sub-task from the task instruction; (the limitation describes devising an initial sub-task by breaking the task instruction down into a smaller task, which is a mental process implemented using a pen and paper.)
generating ... the first sub-task from the initial first sub-task. (the limitation describes refining an initial sub-task into a final sub-task, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
by the first neural network model; by the fourth neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, by the fourth neural network model, the initial first sub-task via the third connection; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 8 and 18,
Step 2A Prong 1: those claims recite the following abstract ideas:
generating ... a first part of the task from the task instruction; (the limitation describes devising a first part of the task from the task instruction, which is a mental process implemented using a pen and paper.)
generating ... a second part of the task from the task instruction; (the limitation describes devising a second part of the task from the task instruction, which is a mental process implemented using a pen and paper.)
assembling ... the first part and the second part of the task instruction to form the first sub-task; (the limitation describes aggregating two parts into a single sub-task, which is a mental process implemented using a pen and paper.)
selecting ... the second neural network model from the plurality of the neural network models based on the first sub-task. (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
by the first neural network model; by the fourth neural network model; by the first neural network model or the second neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 9 and 19,
Step 2A Prong 1: those claims recite the following abstract ideas:
selecting a fifth neural network model from the plurality of the neural network models based on the first sub-task; (the limitation describes selecting which resource is to be assigned the sub-task, which is a mental process implemented in the human mind.)
generate another first sub-task and select a sixth neural network from the plurality of the neural network models based on the other first sub-task; (the limitation describes devising a further sub-task and selecting which resource is to be assigned that sub-task, which is a mental process implemented using a pen and paper.)
generating ... another second sub-task based on the first sub-task and the other first output. (the limitation describes devising a further sub-task by evaluating the sub-task together with the received output, which is a mental process implemented using a pen and paper.)
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
The additional limitation of claim 19 "wherein the operations further comprise" (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
building a fourth connection, via a fourth API, between the first neural network model and the fifth neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
transmitting, by the first neural network model, the first sub-task to the fifth neural network model via the fourth API to cause the fifth neural network model to; (this limitation describes data transmission, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
receiving, via at least the fourth connection, another first output from the sixth neural network model that executes another first sub-task package corresponding to the other first sub-task; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
by the fifth neural network model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 10,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the first neural network and the second neural network are each independently trained. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
Claim Rejections - 35 USC § 103
Claim(s) 1-2, 5-12, and 15-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Siebel et al. (US 20240202225 A1) in view of Cai et al. (US 20230112921 A1).
As to claim 1, Siebel teaches a method for building a hierarchical structure of a plurality of neural network models for performing a task, the method comprising: (see Siebel paragraph [0023] "The enterprise generative artificial intelligence architecture includes an orchestrator agent (or, simply, orchestrator) that supervises, controls, and/or otherwise administrates many different agents and tools. Orchestrators can include one or more machine learning models... Agents can include one or more multimodal models (e.g., large language models) to accomplish the prescribed tasks using a variety of different tools.", and see Siebel paragraph [0031] "the enterprise generative artificial intelligence system architecture and environment includes a hierarchy of layers. More specifically, the hierarchy of layers includes an input layer 202, a supervisory layer 210, an agent layer 220, an agent and tool layer 230, a tool and data model layer 250, and an external layer 280.")
receiving, via a data interface, a task instruction; (see Siebel paragraph [0032] "The input layer 202 represents a layer of the enterprise generative artificial intelligence system architecture that receives an input (e.g., a query, complex input, instruction set, and/or the like) from a user or system. For example, an interface module of the enterprise generative artificial intelligence system may receive the input.")
generating, by a first neural network model, a first sub-task from the task instruction; (see Siebel paragraph [0028] "The orchestrator can use a multimodal model (e.g., large language model) to further process the input 102 to create a plan for determining a result (step 112) for the input. The plan may include a prescribed set of tasks, such as structured data retrieval tasks, unstructured data retrieval tasks, timeseries processing tasks, visualization tasks, and the like.", and see Siebel paragraph [0148] "an orchestrator module (e.g., orchestrator module 504) receives an input (e.g., a complex input) and generates a plan and a corresponding set of prescribed tasks. The orchestrator generates several sub-queries from the input, the plan, and/or the prescribed set of tasks.")
selecting a second neural network model from the plurality of the neural network models based on the first sub-task; (see Siebel paragraph [0023] "Agents can include one or more multimodal models (e.g., large language models) to accomplish the prescribed tasks using a variety of different tools.", and see Siebel paragraph [0029] "the orchestrator may use one or more multimodal models (e.g., language, video, audio, statistical models, etc.), and/or other machine learning models, to interpret the input 102 to select appropriate agents 106 and appropriate tools 108. For example, the orchestrator may determine that a first portion of the input requires a database query, while another portion of the input requires an API call. The orchestrator can appropriately route the first portion of the input to the appropriate agent 106-1 (e.g., a structured data retrieval agent)")
building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model; (see Cai paragraph [0113] "Additionally or alternatively, one or more machine-learned models 190 can be accessed as a service over the network 180. For example, the calls (e.g., requests for inference) can be made to the models 190 using one or more application programming interfaces (APIs).")
generating, by the first neural network model, a first sub-task package in a format compliant with the second neural network model; (see Siebel paragraph [0067] "the orchestrator 504 may transform a first portion of an input into an SQL query and send that to an unstructured data retriever agent module 506-2 agent, transform a second portion of the input into an API call and send that to an API agent module 506-7", and see Siebel paragraph [0227] "first instructions may be determined for a first agent, where the first instructions are determined according to a format and/or language to be received and processed by the first agent (e.g., in dependence on a format of instructions and/or a language which the first agent is configured to receive).")
receiving, via the first connection, a first output from the second neural network model that executes the first sub-task package; (see Siebel paragraph [0141] "The agents 506 and/or tools 508 execute the prescribed set of tasks, and the orchestrator module 504 observes the result.")
generating, by the first neural network model, a second sub-task based on the task instruction and the first output; and (see Siebel paragraph [0141] "The orchestrator module 504 determines whether to submit a final answer or whether the orchestrator module 504 needs more information... Otherwise, the orchestrator module 504 can create another prescribed set of tasks, and the process can continue until the orchestrator module 504 has enough information to answer", and see Siebel paragraph [0224] "The additional retrieval requests may be based on the one or more responses to the prompt.")
causing the task instruction to be jointly performed by one or more selected neural network models from the plurality of neural network models based at least in part on the second sub-task. (see Siebel paragraph [0025] "The different agents can each separately, and in parallel, handle each of these requests, greatly increasing computational efficiency.", and see Siebel paragraph [0141] "the process can continue until the orchestrator module 504 has enough information to answer or if a stopping condition is satisfied (e.g., a maximum number of hops).")
Siebel does not explicitly teach "building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model"
However, Cai teaches
building a first connection, via a first application programming interface (API), between the first neural network model and the second neural network model; (see Cai paragraph [0113] "Additionally or alternatively, one or more machine-learned models 190 can be accessed as a service over the network 180. For example, the calls (e.g., requests for inference) can be made to the models 190 using one or more application programming interfaces (APIs).")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify Siebel to build the connection between the orchestrator model and the selected agent model via an API, as taught by Cai. Both systems coordinate multiple language models to solve a task, and accessing a model through an API call is a known technique that yields the predictable result of a flexible connection between the models, see Cai paragraph [0113].
As to claim 2, Siebel as modified by Cai teaches the method of claim 1,
wherein the causing the task instruction to be jointly performed further comprises: selecting a third neural network model from the plurality of neural network models based on the second sub-task; (see Siebel paragraph [0029] "route the second portion of the input to another agent 106-2 (e.g., API agent). There could be any number of such agents 106 accessing any number of different tools 108.", and see Siebel paragraph [0141] "Otherwise, the orchestrator module 504 can create another prescribed set of tasks, and the process can continue")
generating, by the first neural network model, a second sub-task package in a format compliant with the third neural network model; and (see Siebel paragraph [0227] "Second instructions may be determined for a second agent, where the second instructions are determined according to a format and/or language to be received and processed by the second agent (e.g., in dependence on a format of instructions and/or language which the second agent is configured to receive). The format and/or language of instructions determined for the first agent may be different to the format and/or language of instructions determined for the second agent.")
receiving, via the second connection, a second output from the third neural network model that executes the second sub-task package. (see Siebel paragraph [0222] "receiving, from the one or more agents of the plurality of agents, data from multiple data domains based on instructions from the orchestrator")
Siebel does not explicitly teach "building a second connection, via a second API between the first neural network model and the third neural network model"
However, Cai teaches
building a second connection, via a second API between the first neural network model and the third neural network model; (see Cai paragraph [0113] "the calls (e.g., requests for inference) can be made to the models 190 using one or more application programming interfaces (APIs).")
As to claim 5, Siebel as modified by Cai teaches the method of claim 1,
further comprising: receiving, by one of the first neural network model or the second neural network model, a human instruction; and (see Siebel paragraph [0159] "In step 862, a query is received. For example, the query may be "how much wine do they produce?"... a previous conversation 864 (e.g., as part of a chat with a chat bot) may have included a conversation about France. The system can provide France as contextual information 864 to generate a new query 866, such as "How much wine does France produce?"")
generating, by the first neural network model, the first sub-task based on the task instruction and the human instruction. (see Siebel paragraph [0160] "the enterprise generative artificial intelligence system can generate a rewritten query 866 (e.g., using large language model 865) which can be executed against the vector store 868 to retrieve passages 870.")
As to claim 6, Siebel as modified by Cai teaches the method of claim 1,
further comprising: a fourth neural network model communicatively coupled to the first neural network model via a third connection based on a third API; and (see Cai paragraph [0040] "the present disclosure introduces the concept of chaining instantiations of machine-learned language models (e.g., LLMs) together, where the output of one instantiation becomes the input for the next, and so on, thus aggregating the gains per step.", and see Cai paragraph [0113] "the calls (e.g., requests for inference) can be made to the models 190 using one or more application programming interfaces (APIs).")
generating, by the first neural network or the fourth network, the first sub-task package. (see Cai paragraph [0043] "In a chain, a problem can be broken down into a number of smaller sub-tasks, each mapped to a distinct step with a corresponding prompt; results of one or more previous steps can be aggregated in the next step's input prompt.")
As to claim 7, Siebel as modified by Cai teaches the method of claim 6,
wherein the generating, jointly by the first neural network and the fourth neural network, the first sub-task package comprises: generating, by the first neural network model, an initial first sub-task from the task instruction; (see Cai paragraph [0059] "a LLM chain is used that includes three steps, each for a distinct sub-task: first, a 'split points' step/prompt that extracts each individual presentation problem from the original feedback")
receiving, by the fourth neural network model, the initial first sub-task via the third connection; and (see Cai paragraph [0040] "chaining instantiations of machine-learned language models (e.g., LLMs) together, where the output of one instantiation becomes the input for the next")
generating, by the fourth neural network model, the first sub-task from the initial first sub-task. (see Cai paragraph [0059] "second, an 'ideation' step/prompt that brainstorms suggestions per problem")
As to claim 8, Siebel as modified by Cai teaches the method of claim 6,
selecting, by the first neural network model or the second neural network model, the second neural network model from the plurality of the neural network models based on the first sub-task. (see Siebel paragraph [0029] "the orchestrator may use one or more multimodal models... to interpret the input 102 to select appropriate agents 106 and appropriate tools 108... The orchestrator can appropriately route the first portion of the input to the appropriate agent 106-1 (e.g., a structured data retrieval agent)")
Siebel does not explicitly teach "wherein the generating, jointly by the first neural network and the fourth neural network, the first sub-task package comprises: generating, by the first neural network model, a first part of the task from the task instruction", "generating, by the fourth neural network model, a second part of the task from the task instruction", and "assembling, by the first neural network model or the second neural network model, the first part and the second part of the task instruction to form the first sub-task; and"
wherein the generating, jointly by the first neural network and the fourth neural network, the first sub-task package comprises: generating, by the first neural network model, a first part of the task from the task instruction; (see Cai paragraph [0087] "For example, in the Ideation step (b2), Alex's three presentation problems are addressed in parallel, creating three paths of model calls.")
generating, by the fourth neural network model, a second part of the task from the task instruction; (see Cai paragraph [0087] "Alex's three presentation problems are addressed in parallel, creating three paths of model calls.")
assembling, by the first neural network model or the second neural network model, the first part and the second part of the task instruction to form the first sub-task; and (see Cai paragraph [0087] "But later in Compose Point (b3), the three sets of problems and suggestions are merged into one.")
As to claim 9, Siebel as modified by Cai teaches the method of claim 1,
further comprising: selecting a fifth neural network model from the plurality of the neural network models based on the first sub-task; (see Siebel paragraph [0064] "the orchestrator module 504 may receive a query and instruct agent 506-1 to retrieve associated information.")
transmitting, by the first neural network model, the first sub-task to the fifth neural network model via the fourth API to cause the fifth neural network model to generate another first sub-task and select a sixth neural network from the plurality of the neural network models based on the other first sub-task; (see Siebel paragraph [0064] "The retrieval agent module 506-1 may then select unstructured data retriever agent module 506-2 and/or structured data retriever agent module 506-3 depending on whether the orchestrator module 504 wants to retrieve structured or unstructured data records.", and see Siebel paragraph [0073] "the retrieval agent module 506-1 can coordinate/instruct the unstructured data retriever agent module 506-2 to retrieve unstructured data records")
receiving, via at least the fourth connection, another first output from the sixth neural network model that executes another first sub-task package corresponding to the other first sub-task; and (see Siebel paragraph [0064] "The appropriate agents 506 can the select the corresponding tools and provide the tool output to the orchestrator module 504 and/or comprehension module 510 for determining a final result.")
generating, by the fifth neural network model, another second sub-task based on the first sub-task and the other first output. (see Siebel paragraph [0151] "In step 736, the structured data agent creates a structured data retrieval specification query based on the tool outputs of steps 726-734 and executes that query against a structured datastore.")
Siebel does not explicitly teach "building a fourth connection, via a fourth API, between the first neural network model and the fifth neural network model"
However, Cai teaches
building a fourth connection, via a fourth API, between the first neural network model and the fifth neural network model; (see Cai paragraph [0113] "the calls (e.g., requests for inference) can be made to the models 190 using one or more application programming interfaces (APIs).")
As to claim 10, Siebel as modified by Cai teaches the method of claim 1,
wherein the first neural network and the second neural network are each independently trained. (see Siebel paragraph [0024] "an agent may employ a particular model (e.g., large language model, other machine learning model, and/or data model) that has been trained on industry-specific datasets, such as healthcare datasets.", and see Siebel paragraph [0108] "the comprehension module 510 can use particular models (e.g., data models and/or large language models) for a particular domain (e.g., ... a large language model trained on aerospace-specific datasets) and use another data model and/or large language model for another domain (e.g., ... a large language model trained on defense-specific datasets)")
As to claim 11, this is directed to a system embodiment that corresponds to method claim 1. See the rejection for claim 1 above, which also applies to claim 11. In addition claim 11 recites the additional elements:
a memory that stores the plurality of neural network models and a plurality of processor executable instructions; (see Siebel paragraph [0129] "a model registry 550 can store various models (e.g., machine learning models, large language models, data models) and/or model configurations.", and see Siebel paragraph [0205] "Each of the memory system 1506 and the storage system 1508 comprises a computer-readable medium, which stores instructions or programs executable by processor 1504.")
one or more hardware processors that read and execute the plurality of processor-executable instructions from the memory to perform operations comprising: (see Siebel paragraph [0203] "The computing device 1502 comprises a processor 1504, memory 1506, storage 1508, an input device 1510, a communication network interface 1512, and an output device 1514 communicatively coupled to a communication channel 1516. The processor 1504 is configured to execute executable instructions (e.g., programs). In some embodiments, the processor 1504 comprises circuitry or any processor capable of processing the executable instructions.")
As to claim 12, this is directed to a system embodiment that corresponds to method claim 2. See the rejection for claim 2 above, which also applies to claim 12.
As to claim 15, this is directed to a system embodiment that corresponds to method claim 5. See the rejection for claim 5 above, which also applies to claim 15.
As to claim 16, this is directed to a system embodiment that corresponds to method claim 6. See the rejection for claim 6 above, which also applies to claim 16.
As to claim 17, this is directed to a system embodiment that corresponds to method claim 7. See the rejection for claim 7 above, which also applies to claim 17.
As to claim 18, this is directed to a system embodiment that corresponds to method claim 8. See the rejection for claim 8 above, which also applies to claim 18.
As to claim 19, this is directed to a system embodiment that corresponds to method claim 9. See the rejection for claim 9 above, which also applies to claim 19.
As to claim 20, this is directed to a computer-program embodiment that corresponds to method claim 1. See the rejection for claim 1 above, which also applies to claim 20. In addition claim 11 recites the additional elements:
A non-transitory machine-readable medium comprising a plurality of machine-executable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform operations comprising (see Siebel paragraph [0205] "Each of the memory system 1506 and the storage system 1508 comprises a computer-readable medium, which stores instructions or programs executable by processor 1504.", and see Siebel paragraph [0247] "According to examples disclosed herein there is provided a non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to perform: processing, by an orchestrator, an input; selecting, by the orchestrator based on the processed input, a first agent of a plurality of different agents")
Claim(s) 3-4, and 13-14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Siebel et al. (US 20240202225 A1) in view of Cai et al. (US 20230112921 A1) and Medford et al. (US 20250165890 A1).
As to claim 3, Siebel as modified by Cai teaches the method of claim 1,
Siebel as modified by Cai does not explicitly teach "wherein the first sub-task package comprises a first prompt compliant with the second neural network model, instructing the second neural network model to perform the first sub-task"
However, Medford teaches
wherein the first sub-task package comprises a first prompt compliant with the second neural network model, instructing the second neural network model to perform the first sub-task. (see Medford paragraph [0050] "When the Chat Manager Agent 212 is tasked with generating inputs, such as LLM prompts for the various specialized agents, it relies on the MemGPT Agent 104 to provide the necessary context.", and see Medford paragraph [0051] "This enriched context allows the Chat Manager Agent 212 to formulate a precise and informed prompt for the Engineer Agent 208, who then uses this information to effectively modify or enhance the software module.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify Siebel as modified by Cai to include a prompt in the sub-task package, as taught by Medford. Both systems delegate sub-tasks to LLM-based agents, and using a prompt tailored to the receiving agent is a known technique that yields the predictable result of the agent directly acting on the delegated sub-task, see Medford paragraph [0051].
As to claim 4, Siebel as modified by Cai teaches the method of claim 1,
Siebel as modified by Cai does not explicitly teach "wherein the first output comprises one or more of a completion status corresponding to the first sub-task, identification information for the first neural network model, or identification information for the second neural network model"
However, Medford teaches
wherein the first output comprises one or more of a completion status corresponding to the first sub-task, identification information for the first neural network model, or identification information for the second neural network model. (see Medford paragraph [0069] "Once the code is executed, the Executor Agent 210 collects data on its performance, including execution logs, error reports, and output results. This data provides tangible feedback on the code's real-world operability. The Chat Manager Agent 212 then processes this output to determine the success of the execution.", and see Medford paragraph [0070] "the results, including any execution logs, error reports, and performance data, are passed back to the Engineer Agent 208 via the Chat Manager Agent 212.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify Siebel as modified by Cai to include a completion status in the returned output, as taught by Medford. Both systems return agent results to a managing agent, and returning feedback indicating the success of the execution is a known technique that yields the predictable result of the orchestrator knowing whether the sub-task was completed before assigning the next one, see Medford paragraph [0069].
As to claim 13, this is directed to a system embodiment that corresponds to method claim 3. See the rejection for claim 3 above, which also applies to claim 13.
As to claim 14, this is directed to a system embodiment that corresponds to method claim 4. See the rejection for claim 4 above, which also applies to claim 14.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABDULLAH K ABOUD whose telephone number is (571)272-0025. The examiner can normally be reached Mon-Fri 8am-5pm.
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, Li B Zhen, can be reached at (571) 272-3768. 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.
/ABDULLAH KHALED ABOUD/ Examiner, Art Unit 2121
/Li B. Zhen/ Supervisory Patent Examiner, Art Unit 2121
Prosecution Insights