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 .
Response to Arguments
Applicant's arguments filed 5/6/26 have been fully considered but they are not persuasive.
Regarding the 35 U.S.C. 101 rejection of the claims, Applicant argues that the claims include limitations “determining, by the first LLM using the first prompt data, a first recognized entity in the first natural language request” and “generating, by the first LLM using the second prompt data, first instructions to perform the first recommendation action using the first content identifier and second content identifier” that do not merely recite mental processes and that “generating constrained LLM outputs by enforcing deterministic guard rails on probabilistic LM inference” has no analogy to mental processes, and as such, argues that that the claims do not recite mental processes (Arguments, pg. 11, third para. -pg. 12, second para.).
Examiner respectfully disagrees as steps “determining, by the first LLM using the first prompt data, a first recognized entity in the first natural language request” and “generating, by the first LLM using the second prompt data, first instructions to perform the first recommendation action using the first content identifier and second content identifier” correspond to using LLM to analyze/evaluate data, i.e. tying a computer component (see Applicant’s original specification para. [0097] and [00110]) to analysis/evaluation steps, and as such, corresponds to mental processes. Also, “generating constrained LLM outputs by enforcing deterministic guard rails on probabilistic LM inference” corresponds to unclaimed language.
Regarding arguments presented pg. 12, third para. – pg. 17, second para., Examiner respectfully disagrees that paras. [0012]-[0015] provide evidence of technical improvements arising from LM hallucinations, as reducing hallucinations (i.e., errors in provided outputs/answers) does not provide any improvement to other technology or to a computer implementing the claimed steps, where the reduction is an improvement to the abstract idea, and a benefit that flows from performing an abstract idea in conjunction with a generic computer, and where automating an abstract idea to conserve human resources and minimize errors is a 'quintessential do it on a computer patent' directed to an abstract idea" (see Univ. of Fla. Research Found., Inc. V. Gen. Elec. Co., 916 F.3d 1363, 1367 (Fed. Cir. 2019).
Regarding arguments presented pg. 17, third para. – pg. 18, third para., Examiner respectfully disagrees as solving Hallucination simply by delegating functioning according to limitations "determining, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system” and "routing, by the generalist NLP system, the first query data to the constrained NLP system," does not provide an inventive concepts as delegating tasks from a generic system to a specialized system is well known as presented by the teachings of Karri (see rejection below). Furthermore, determining (i.e., analyzing or evaluating) data and routing (i.e., transmitting data) have been identified by the courts as corresponding to well understood, routine, and convention activities as provided in the MPEP (2106.05(d)II). Therefore, Examiner maintains the rejection of the claims.
Regarding the 35 U.S.C. 103 rejection of claims 5 and 13 with references Karri and Miller, the rejection of similar claim 1 with Miller in view of Krishnan as well as claims dependent therefrom, Applicant argues that para. [0027] of Karri merely describes common AI service classifying a query and generating an agent change to execute the query but does not contemplate routing a request from a generalist system to a constrained system with a restricted execution model, and as such argues that Karr fails to disclose limitations "determining, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system," "routing, by the generalist NLP system, the first query data to the constrained NLP system," or "determining, by the constrained NLP system and using the first query data, first action data associated with the first query data," (Arguments, pg. 18, fourth para. – pg. 20, third para.).
Examiner respectfully disagrees as Karri discloses using its common AI service 110 (i.e., the claimed generalist NLP system) interacting with an agent library 125 (i.e., the claimed constrained NLP system) that contains specialized agents, where the common AI service 110 invokes the specialized agents in the agent library 125 to perform various actions to orchestrate workflows in response to a user’s request (para. [0055]), where the specialized agents include an entity resolution agent 310 that interacts with a vector database 115 for entity mapping, corresponding to a determination that the user’s query is delegated to a specialized agent of the agent library 125 by a common AI service 110, and as such, limitation "determining, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system," and “routing, by the generalist NLP system, the first query data to the constrained NLP system”. Karri also discloses the entity resolution agent 310 issuing queries to search for semantically similar queries stored in the vector database 115 compared to a user’s query (para. [0066]), corresponding to “determining, by the constrained NLP system and using the first query data, first action data associated with the first query data;”. Therefore, Examiner maintains the rejection of claim 5 as well as the rejection of similar independent clams 1 and 15, and the claims dependent therefrom.
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.
1. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to the abstract idea of query/input analysis without significantly more. The claim 1 recites steps of receiving first query data comprising a first natural language request to recommend content of a first type/related to content (i.e., a data gathering step), determining that the query data is to be processed by a constrained system (i.e., a data analysis/judgement step), routing he first query data to a constrained system (i.e. a data analysis/transmission step),determining first action data associated with the first query (i.e. a data analysis/evaluation step), performing a lookup using the first query data to determine a first action identifier/data associated with a first recommendation action/query data (i.e., a data evaluation/analysis step), generating, by a large language model (LLM) orchestrator, first prompt data comprising the first query data and the first recommendation action, wherein the first prompt data instructs a first LLM to recognize entities in the first query data relevant to the first recommendation action (i.e., a data evaluation/analysis step), determining, by the first LLM using the first prompt data, a first recognized entity in the first natural language request, wherein the first recognized entity corresponds to the content of the first type (i.e., a data evaluation/analysis step), sending, by the first LLM to the LLM orchestrator, a request to resolve the first recognized entity (i.e., a data evaluation/analysis step), determining, using a first keyword resolver tool, a first content identifier related to a first instance of the content of the first type and a second content identifier related to a second instance of the content of the first type (i.e., a data evaluation/analysis/judgement step), generating, by the LLM orchestrator, second prompt data comprising the first content identifier and the second content identifier (i.e., a data evaluation/analysis step), generating, by the first LLM using the second prompt data, first instructions to perform the first recommendation action using the first content identifier and the second content identifier (i.e., a data evaluation/analysis step) receiving, by an action execution component, the first instructions (i.e., a data evaluation/analysis step) and executing the first recommendation action by the action execution component, wherein the executing the first recommendation action comprises causes an output of a list comprising data representing the first instance of the content of the first type and data representing the second instance of the content of the first type (i.e., a post solutional step of outputting data as a result of analysis), corresponding to steps achievable by a human in manually/mentally analyzing data to determine output to provide, and as such, the steps correspond to the mental processes category of abstract ideas.
The claims 5 and 13 recite steps of: receiving first query data comprising a first request related to first content i.e., a data gathering step), determining based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system (i.e., a data evaluation/analysis step), routing the first query data to the constrained NLP system (i.e., a data evaluation/transmission step), determining using the first query data, first action data associated with the first query data (i.e., a data evaluation/analysis step), generating first prompt data comprising a representation of the first query data and data representing the first action data, wherein the first prompt data instructs a first language model (LM) of the constrained NLP system to recognize entities in the first query data relevant to the first action data (i.e., a data evaluation/analysis step), determining, by the first LM based at least in part on the first prompt data, a first recognized entity from the first request, wherein the first recognized entity is associated with the first content (i.e., a data evaluation/analysis step), generating a request to resolve the first recognized entity (i.e., a data evaluation/analysis step), determining a first resolved entity of the constrained NLP system for the first recognized entity (i.e., a data evaluation/analysis step), generating second prompt data comprising the first resolved entity (i.e., a data evaluation/analysis step), generating based at least in part on the second prompt data, first instructions to perform at least one action associated with the first action data using the first resolved entity (i.e., a data evaluation/analysis step), and generating output data associated with the first resolved entity based at least in part on the first instructions (i.e., a data evaluation/analysis step), corresponding to steps achievable by a human in manually/mentally analyzing data to determine output to provide, and as such, the steps correspond to the mental processes category of abstract ideas.
This judicial exception is not integrated into a practical application because the claims are directed to an abstract idea with additional generic computer elements, where the generically recited computer elements (computer-implemented, LLM, LM, system, generalist NLP system, constrained NLP system, LLM orchestrator, processor, memory) do not add a meaningful limitation to the abstract idea because they amount to simply implementing the abstract idea on a computer (see paras. [0097] and [00110] of Applicant’s original specification). The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because steps “generating, by the first LM based at least in part on the second prompt data, first instructions to perform at least one action associated with the first action data using the first resolved entity”, generating output data associated with the first resolved entity based at least in part on the first instructions”, and “executing the first recommendation action by the action execution component, wherein the executing the first recommendation action comprises causes an output of a list comprising data representing the first instance of the content of the first type and data representing the second instance of the content of the first type”, correspond to the well-understood, routine, conventional computer functions of “Gathering and analyzing information using conventional techniques and displaying the result,” and “collecting information, analyzing it, and displaying certain results of the collection and analysis” as recognized by the court decisions listed in MPEP § 2106.05 and as provided by cited references Karri and Krishnan (PTO 892 form).
The dependent claims also recite mental processes and do not add significantly more than the abstract idea and are as such similarly rejected.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
2. Claims 1-20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. In particular, amended claims 1, 5 and 13 recite the use of “a constrained natural language processing (constrained NLP) system”. There is no disclosure of this amended language as part of Applicant’s original disclosure. The portions of the specification (para. [0015]-[0019]) that Applicant points to as providing support for the amendments to the claims (Remarks, 5/5/26, pg. 11, first para.) describe the use of specialized/specialist models or domain specific systems, but not constrained systems. A constrained system working according to strict rules/constraints is not the same as a specialized model working to handle a single/specific task.
The dependent claims are rejected based on their dependency.
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.
3. Claims 12 and 20 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.
In particular, claims 12 and 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being incomplete for omitting essential elements, such omission amounting to a gap between the elements. See MPEP § 2172.01. Independent claims 5 and 13 from which the claims depend from recite “a first Language model (LM) of the constrained NLP system.
The omitted elements are: “a first LM of the generalist NLP system” before the claimed “second LM of the generalist system” claimed in claims 12 and 20.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
4. Claims 5, 6, 12-14 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Karri et al US 2025/0077263 A1 (“Karri”) in view of Miller et al US 2024/0289545 A1 (“Miller”)
Per claim 5, Karri discloses a method comprising:
receiving, by a generalist natural language processing (generalist NLP) system, first query data comprising a first request related to first content (fig. 3, element 110; The system 100 includes a conversational planning user interface 105 for receiving natural language input from end users; a common artificial intelligence (AI) service 110 for processing natural language input …, para. [0035]; para. [0066]);
determining, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system (The common AI service 110 interacts with the agent library 125 that contains specialized agents including, but not limited to an entity resolution agent 310 that extracts and maps entities from user 305 requests … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
routing, by the generalist NLP system, the first query data to the constrained NLP system (fig. 3; an agent library 125 containing various specialized agents …, para. [0035]; The common AI service 110 orchestrates interactions between agents in the agent library 125 and a user 305 based on requests submitted by the user 305 via the conversational planning user interface 105. The common AI service 110 interacts with the agent library 125 that contains specialized agents … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
determining, by the constrained NLP system and using the first query data, first action data associated with the first query data (fig. 6, element 615; At operation 505, a user data input is received as a natural language request entered via a user interface (e.g., the conversational planning user interface 105 as described in FIG. 1, etc.). At operation 410 a common AI service (e.g., the common AI service 110 as described in FIG. 1, etc.) initiates intent classification of the data input to identify an update intent…. At operation 520 intents with a probability within a threshold (e.g., top-k intents, etc. are retrieved based on results from the vector database query to generate an intent candidate set…., para. [0066]);
generating first prompt data comprising a representation of the first query data, wherein the first prompt data instructs a first language model (LM) of the constrained NLP system to recognize entities in the first query data relevant to the first action data (fig. 6, elements 615, 620; para. [0039]; At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent…. by way of example and not limitation, an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; At operation 810 an entity resolution agent (e.g., the entity resolution agent 310 as described in FIG. 3, etc.) initiates an entity extraction and mapping process. At operation 815, potential entities are identified in the user request.…, para. [0079]);
determining, by the first LM based at least in part on the first prompt data, a first recognized entity from the first request, wherein the first recognized entity is associated with the first content (At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent. The orchestrator selects the agents from an agent library … an entity resolution agent that extracts and maps entities from user utterances …, para. [0050]; para. [0066]; para. [0079]; para. [0088]-[0089]);
generating, by the first LM, a request to resolve the first recognized entity (fig. 3, element 310; fig. 8, elements 840, 845; an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; para. [0060]; At operation 815, potential entities are identified in the user request.… if multiple results are returned for a specific phrase, options are provided to the user interface for user disambiguation. At operation 845, matched entities are mapped to corresponding system entities.…, para. [0079]);
determining a first resolved entity of the constrained NLP system for the first recognized entity (fig. 3; para. [0050]; para. [0060]; para. [0070]; At operation 815, potential entities are identified in the user request.… if multiple results are returned for a specific phrase, options are provided to the user interface for user disambiguation. At operation 845, matched entities are mapped to corresponding system entities.…, para. [0079])
generating second prompt data comprising the first resolved entity (para. [0023]; para. [0050]-[0051]; para. [0057]; an update agent (e.g., the update agent 315 as described in FIG. 3, etc.) creates a structured query based on the identified intent and resolved entities …, para. [0060]; para. [0080]-[0081]);
generating, by the first LM based at least in part on the second prompt data, first instructions to perform at least one action associated with the first action data using the first resolved entity (an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.), an update agent that creates update queries based on user input and entities …, para. [0050]-[0051]; para. [0053]; para. [0060]-[0063]); and
generating output data associated with the first resolved entity based at least in part on the first instructions (At operation 230, results generated in the virtual container are output to the user interface …, para. [0052]; para. [0060]-[0063])
Karri does not explicitly disclose generating first prompt data comprising data representing the first action data
However, this feature is taught by Miller (para. [0079]; a solution plan (e.g., 156) may be generated based on the goal (e.g., 154) and the user prompt (112) …, para. [0080]; the solution plan (e.g., 156) that was generated by the plan creation component (e.g., 150) may be passed to the large language mode (e.g., 130) for further processing. …, para. [0082])
It would have been obvious to one of ordinary skill in the art before the effective filing of the instant invention to combine the teachings of Miller with the method of Karri in arriving at the missing features of Karri, because such combination would have resulted in enhancing the quality and accuracy of solution plans (Miller, para. [0174]).
Per claim 6, Karri in view of Miller discloses the method of claim 5,
Karri discloses performing, by the constrained NLP system, a lookup of a first data store using the first query data to retrieve the first action data, wherein the first action data comprises a set of predefined tasks defined for compliance with a set of rules (fig. 3; para. [0035]; para. [0066]; The orchestrator 155 integrates the composite actions into an overall workflow for execution of tasks for intents detected from user requests …, para. [0076]).
Per claim 12, Karri in view of Miller discloses the method of claim 5,
Karri discloses: determining, by a second LM of the generalist NLP system, a first domain associated with the first query data (fig. 3; fig. 7; para. [0037]-[0039]); and
routing by the generalist NLP system, the first query data to the constrained NLP system based on the first domain (fig. 3; fig. 7; para. [0020]; para. [0037]-[0039]; para. [0055]; para. [0066]).
Per claim 13, Karri discloses a system comprising:
at least one processor (para. [0095]-[0096]), and
non-transitory computer-readable memory storing instructions that, when executed by the at least one processor (para. [0095]-[0096]), are effective to:
receive, by a generalist natural language processing (generalist NLP) system, first query data comprising a first request related to first content (fig. 3, element 110; The system 100 includes a conversational planning user interface 105 for receiving natural language input from end users; a common artificial intelligence (AI) service 110 for processing natural language input …, para. [0035]; para. [0066]);
determine, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system (The common AI service 110 interacts with the agent library 125 that contains specialized agents including, but not limited to an entity resolution agent 310 that extracts and maps entities from user 305 requests … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
route, by the generalist NLP system, the first query data to the constrained NLP system (fig. 3; an agent library 125 containing various specialized agents …, para. [0035]; The common AI service 110 orchestrates interactions between agents in the agent library 125 and a user 305 based on requests submitted by the user 305 via the conversational planning user interface 105. The common AI service 110 interacts with the agent library 125 that contains specialized agents … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
determine, by the constrained NLP system, using the first query data, first action data associated with the first query data (fig. 3; fig. 5, elements 505, 520; para. [0055]-[0056]; At operation 520 intents with a probability within a threshold (e.g., top-k intents, etc. are retrieved based on results from the vector database query to generate an intent candidate set.…, para. [0066]);
generate first prompt data comprising a representation of the first query data, wherein the first prompt data instructs a first language model (LM) of the constrained NLP system to recognize entities in the first query data relevant to the first action data (fig. 6, elements 615, 620; para. [0039]; At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent…. by way of example and not limitation, an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; At operation 810 an entity resolution agent (e.g., the entity resolution agent 310 as described in FIG. 3, etc.) initiates an entity extraction and mapping process. At operation 815, potential entities are identified in the user request.…, para. [0079]);
determine, by the first LM based at least in part on the first prompt data, a first recognized entity from the first request, wherein the first recognized entity is associated with the first content (At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent. The orchestrator selects the agents from an agent library … an entity resolution agent that extracts and maps entities from user utterances …, para. [0050]; para. [0066]; para. [0079]; para. [0088]-[0089]);
generate a request to resolve the first recognized entity (fig. 3, element 310; fig. 8, elements 840, 845; an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; para. [0060]);
determine a first resolved entity of the constrained NLP system for the first recognized entity (para. [0050]; para. [0060]; para. [0070]; At operation 845, matched entities are mapped to corresponding system entities.…, para. [0079])
generate second prompt data comprising the first resolved entity (para. [0023]; para. [0050]-[0051]; para. [0057]; an update agent (e.g., the update agent 315 as described in FIG. 3, etc.) creates a structured query based on the identified intent and resolved entities …, para. [0060]; para. [0080]-[0081]);
generate, by the first LM based at least in part on the second prompt data, first instructions to perform at least one action associated with the first action data using the first resolved entity (an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.), an update agent that creates update queries based on user input and entities …, para. [0050]-[0051]; para. [0053]; para. [0060]-[0063]); and
generate output data associated with the first resolved entity based at least in part on the first instructions (At operation 230, results generated in the virtual container are output to the user interface …, para. [0052]; para. [0060]-[0063])
Karri does not explicitly disclose generating first prompt data comprising data representing the first action data
However, this feature is taught by Miller (para. [0079]; a solution plan (e.g., 156) may be generated based on the goal (e.g., 154) and the user prompt (112) …, para. [0080]; the solution plan (e.g., 156) that was generated by the plan creation component (e.g., 150) may be passed to the large language mode (e.g., 130) for further processing. …, para. [0082])
It would have been obvious to one of ordinary skill in the art before the effective filing of the instant invention to combine the teachings of Miller with the method of Karri in arriving at the missing features of Karri, because such combination would have resulted in enhancing the quality and accuracy of solution plans (Miller, para. [0174]).
Per claim 14, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 14 and method claim 6 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 14 is similarly rejected under the same rationale as applied above with respect to claim 6.
Per claim 20, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 20 and method claim 12 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 20 is similarly rejected under the same rationale as applied above with respect to claim 12.
5. Claims 1-4, 7-11 and 15-19 are rejected under 35 U.S.C. 103 as being unpatentable over Karri in view of Miller and Krishnan et al US 2022/0093101 A1 (“Krishnan”)
Per claim 1, Karri discloses a computer-implemented method comprising:
receiving, by a generalist natural language processing (generalist NLP) system, first query data comprising a first natural language request (fig. 3, element 110; The system 100 includes a conversational planning user interface 105 for receiving natural language input from end users; a common artificial intelligence (AI) service 110 for processing natural language input …, para. [0035]; para. [0066]);
determining, by the generalist NLP system and based on the first query data, that the first query data is to be processed by a constrained natural language processing (constrained NLP) system (The common AI service 110 interacts with the agent library 125 that contains specialized agents including, but not limited to an entity resolution agent 310 that extracts and maps entities from user 305 requests … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
routing, by the generalist NLP system, the first query data to the constrained NLP system (fig. 3; an agent library 125 containing various specialized agents …, para. [0035]; The common AI service 110 orchestrates interactions between agents in the agent library 125 and a user 305 based on requests submitted by the user 305 via the conversational planning user interface 105. The common AI service 110 interacts with the agent library 125 that contains specialized agents … The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]);
performing, by the constrained NLP system, a lookup using the first query data to determine a first action identifier associated with a first recommendation action (fig. 3; fig. 5; fig. 6; para. [0050]; The common AI service 110 may invoke the agents to perform various actions to orchestrate workflows. For example, the entity resolution agent 310 interacts with the vector database 115 for entity mapping, para. [0055]-[0056]; At operation 515 an entity resolution agent (e.g., the entity resolution agent 310 as described in FIG. 3, etc.) issues queries to search for semantically similar queries stored in a vector database … At operation 520 intents with a probability within a threshold (e.g., top-k intents, etc. are retrieved based on results from the vector database query to generate an intent candidate set.…, para. [0066]);
generating, by a large language model (LLM) orchestrator, first prompt data comprising the first query data, wherein the first prompt data instructs a first LLM of the constrained NLP system to recognize entities in the first query data relevant to the first recommendation action (fig. 3; fig. 6, elements 615, 620; para. [0039]; At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent…. by way of example and not limitation, an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; At operation 810 an entity resolution agent (e.g., the entity resolution agent 310 as described in FIG. 3, etc.) initiates an entity extraction and mapping process. At operation 815, potential entities are identified in the user request.…, para. [0079]);
determining, by the first LLM using the first prompt data, a first recognized entity in the first natural language request (At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent. The orchestrator selects the agents from an agent library … an entity resolution agent that extracts and maps entities from user utterances …, para. [0050]; para. [0066]; para. [0079]; para. [0088]-[0089]);
sending, by the first LLM to the LLM orchestrator, a request to resolve the first recognized entity (fig. 3, element 310; fig. 8, elements 840, 845; an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; para. [0060]);
generating, by the LLM orchestrator, second prompt data (para. [0023]; para. [0050]-[0051]; para. [0057]; an update agent (e.g., the update agent 315 as described in FIG. 3, etc.) creates a structured query based on the identified intent and resolved entities …, para. [0060]; para. [0080]-[0081]);
generating, by the first LLM using the second prompt data, first instructions (para. [0023]; para. [0050]-[0051]; para. [0057]; an update agent (e.g., the update agent 315 as described in FIG. 3, etc.) creates a structured query based on the identified intent and resolved entities …, para. [0060]; para. [0080]-[0081]);
Karri does not explicitly disclose generating first prompt data comprising the first recommendation action
However, this feature is taught by Miller (para. [0079]; a solution plan (e.g., 156) may be generated based on the goal (e.g., 154) and the user prompt (112) …, para. [0080]; the solution plan (e.g., 156) that was generated by the plan creation component (e.g., 150) may be passed to the large language mode (e.g., 130) for further processing. …, para. [0082])
Karri in view of Miller does not explicitly disclose receiving first query data comprising a first natural language request to recommend content of a first type, wherein the first recognized entity corresponds to the content of the first type, generating, second prompt data comprising the first content identifier and the second content identifier, generating, using the second prompt data, first instructions to perform the first recommendation action using the first content identifier and the second content identifier, receiving, by an action execution component, the first instructions or executing the first recommendation action by the action execution component, wherein the executing the first recommendation action comprises causes an output of a list comprising data representing the first instance of the content of the first type and data representing the second instance of the content of the first type.
However, these features are taught by Krishnan:
receiving first query data comprising a first natural language request to recommend content of a first type (fig. 14, element 1404; para. [0048]; para. [0058]; In the example “play songs by the stones,” …, para. [0225]; “tell me a recipe for pasta sauce,” …, para. [0254])
wherein the first recognized entity corresponds to the content of the first type (para. [0058]; In this manner, the NER component 862 identifies “slots” (each corresponding to one or more particular words in text data) that may be useful for later processing. The NER component 862 may also label each slot with a type …, para. [0186]-[0188]; para. [0193]; para. [0225]; para. [0254]);
determining, using a first keyword resolver tool of the constrained NLP system, a first content identifier related to a first instance of the content of the first type and a second content identifier related to a second instance of the content of the first type (para. [0186]-[0188]; para. [0225]; para. [0254]);
generating, second prompt data comprising the first content identifier and the second content identifier (fig. 9; para. [0074]; para. [0254]);
generating, using the second prompt data, first instructions to perform the first recommendation action using the first content identifier and the second content identifier (fig. 9; para. [0074]; para. [0254]; para. [0356]);
receiving, by an action execution component, the first instructions (fig. 2; para. [0069]; para. [0074]; The action selector 1118 determines an action to be performed in response to the user request …, para. [0303]-[0304]); and
executing the first recommendation action by the action execution component, wherein the executing the first recommendation action comprises causes an output of a list comprising data representing the first instance of the content of the first type and data representing the second instance of the content of the first type (fig. 14; fig. 15 F; para. [0074]; para. [0225]; para. [0303]-[0304]; para. [0337]; in response to a user command or otherwise, system 120 may send to device 110 a directive to output audio TTS audio data 1302 corresponding to a list of choices. The list may correspond to a variety of choices such as items to be selected, actions to be selected, or other examples …, para. [0356])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Miller with the method of Karri in arriving at the missing features of Karri, as well as to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in enhancing the quality and accuracy of solution plans (Miller, para. [0174]), as well as in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 2, Karri in view of Miller and Krishnan discloses the computer-implemented method of claim 1,
Krishnan discloses: searching, by the first keyword resolver tool using the first recognized entity, first historical context data comprising a prior input natural language request and a prior response to the prior input natural language request (fig. 15F; para. [0282]; para. [0355]; para. [0387]-[0388]) and
determining, by the first keyword resolver tool, that the prior response comprises the first instance of the content of the first type and the second instance of the content of the first type, wherein the determining the first content identifier and the second content identifier comprises resolving the first recognized entity using the first historical context data (para. [0387]-[0388]).
Per claim 3, Karri in view of Miller and Krishnan discloses the computer-implemented method of claim 1,
Karri discloses receiving second query data comprising a second natural language request (fig. 2; para. [0053]);
performing, by the constrained NLP system a lookup using the second query data to determine a second action identifier (fig. 3; fig. 5, elements 505, 520; para. [0053]; para. [0055]-[0056]; At operation 520 intents with a probability within a threshold (e.g., top-k intents, etc. are retrieved based on results from the vector database query to generate an intent candidate set.…, para. [0066]);
generating, by the LLM orchestrator, third prompt data comprising the second query data, wherein the third prompt data instructs the first LLM to recognize entities in the second query data (fig. 2; para. [0039]; At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent…. by way of example and not limitation, an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; para. [0053]; para. [0066]; para. [0070]; para. [0079])
determining, by the first LLM using the third prompt data, a second recognized entity in the second natural language request (fig. 2; para. [0040]; At operation 220, an orchestrator (e.g., the orchestrator 155 as described in FIG. 1, etc.) selects and executes appropriate agents based on an identified intent. The orchestrator selects the agents from an agent library … an entity resolution agent that extracts and maps entities from user utterances …, para. [0050]; para. [0053]; para. [0066]; para. [0088]-[0089])
sending, by the LLM to the LLM orchestrator, a request to resolve the second recognized entity (fig. 3, element 310; fig. 8, elements 840, 845; an entity resolution agent that extracts and maps entities from user utterances (e.g., requests, etc.) …, para. [0050]; para. [0060]);
Krishnan discloses: receiving second query data comprising a second natural language request that requests a display of content in the list, wherein the second natural language request identifies the content in the list using an ordinal reference (fig. 15F; para. [0073]; para. [0352]);
performing a lookup using the second query data to determine a second action identifier associated with a display action (fig. 15F; para. [0041]; para. [0073]; para. [0189]; para. [0352]; para. [0390]);
generating, third prompt data comprising the display action, wherein the third prompt data instructs the first LLM to recognize entities in the second query data relevant to the display action (para. [0337]; para. [0352]-[0353]; para. [0363]; para. [0387], NLU as suggesting LLM);
determining, using the third prompt data, a second recognized entity in the second natural language request, wherein the second recognized entity corresponds to the content in the list (para. [0041]; para. [0352]; para. [0363]);
determining, using an ordinal resolver tool of the constrained NLP system, the second content identifier based on an order of output of the second instance of the content of the first type in the list corresponding to the ordinal reference (para. [0088]; para. [0352]; para. [0362]-[0364]).
Per claim 4, Karri in view of Miller and Krishnan discloses the computer-implemented method of claim 1,
Karri discloses: determining, by a second LLM of the generalist NLP system, a first domain associated with the first query data (fig. 3; fig. 7; para. [0037]-[0039]); and
routing by the generalist NLP system, the first query data to the constrained NLP system based on the first domain (fig. 3; fig. 7; para. [0020]; para. [0037]-[0039]; para. [0055]; para. [0066]).
Per claim 7, Karri in view of Miller discloses the method of claim 5,
Karri in view of Miller does not explicitly disclose determining, using the first query data, a target for the first action data, wherein the target specifies a predefined data structure that is acted upon using the first action data
However, this feature is taught by Krishnan (para. [0177]; para. [0190]; para. [0225]).
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 8, Karri in view of Miller discloses the method of claim 5, further comprising:
Karri in view of Miller does not explicitly disclose determining, using the first query data, an endpoint for the first action data, wherein the endpoint defines a device for outputting the output data or sending the output data to the endpoint
However, these features are taught by Krishnan:
determining, using the first query data, an endpoint for the first action data, wherein the endpoint defines a device for outputting the output data (para. [0210]-[0213]); and
sending the output data to the endpoint (para. [0210]-[0213])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 9, Karri in view of Miller discloses the method of claim 5,
Karri in view of Miller does not explicitly disclose sending data representing the first recognized entity to an entity information retrieval component of the constrained NLP system, determining, by the entity information retrieval component, at least a first keyword of the first recognized entity, determining an account associated with the first query data or determining the first resolved entity by searching a list or history associated with the account, wherein the first resolved entity comprises identifier data identifying an instance of the first content
However, these features are taught by Krishnan:
sending data representing the first recognized entity to an entity information retrieval component of the constrained NLP system (para. [0088]; para. [0184]);
determining, by the entity information retrieval component, at least a first keyword of the first recognized entity (para. [0184]);
determining an account associated with the first query data (Thus the system may respond to each user with personalized content by keeping track of what each user said in the dialog history…., para. [0355]); and
determining the first resolved entity by searching a list or history associated with the account, wherein the first resolved entity comprises identifier data identifying an instance of the first content (para. [0355])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 10, Karri in view of Miller discloses the method of claim 5,
Karri in view of Miller does not explicitly disclose determining that the first request comprises an ordinal reference or determining the first resolved entity by searching a list of content output prior to receiving the first query data using the ordinal reference, wherein the first resolved entity comprises identifier data identifying an instance of the first content.
However, these features are taught by Krishnan:
determining that the first request comprises an ordinal reference (para. [0352]); and
determining the first resolved entity by searching a list of content output prior to receiving the first query data using the ordinal reference, wherein the first resolved entity comprises identifier data identifying an instance of the first content (para. [0352])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 11, Karri in view of Miller discloses the method of claim 5,
Karri does not explicitly disclose sending data representing the first recognized entity to an entity information retrieval component of the constrained NLP system, determining, by the entity information retrieval component, at least a first keyword of the first recognized entity, searching, using at least the first keyword, first historical context data comprising a prior input request and a prior response to the prior input request or determining the first resolved entity based at least in part on a correspondence between at least the first keyword and a previously-resolved entity present in the prior input request and the prior response
However, these features are taught by Krishnan:
sending data representing the first recognized entity to an entity information retrieval component of the constrained NLP system (para. [0088]; para. [0184]);
determining, by the entity information retrieval component, at least a first keyword of the first recognized entity (para. [0184]);
searching, using at least the first keyword, first historical context data comprising a prior input request and a prior response to the prior input request (para. [0283]; para. [0355]); and
determining the first resolved entity based at least in part on a correspondence between at least the first keyword and a previously-resolved entity present in the prior input request and the prior response (para. [0283]; para. [0355]);
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Krishnan with the method of Karri in view of Miller in arriving at the missing features of Karri in view of Miller, because such combination would have resulted in improving the likelihood that a speech recognition system will output hypotheses that make sense grammatically (Krishnan, para. [0041]; para. [0166]).
Per claim 15, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 15 and method claim7 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 15 is similarly rejected under the same rationale as applied above with respect to claim 7.
Per claim 16, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 16 and method claim 8 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 16 is similarly rejected under the same rationale as applied above with respect to claim 8.
Per claim 17, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 17 and method claim 9 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 17 is similarly rejected under the same rationale as applied above with respect to claim 9.
Per claim 18, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 18 and method claim 10 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 18 is similarly rejected under the same rationale as applied above with respect to claim 10.
Per claim 19, Karri in view of Miller discloses the system of claim 13, and the non-transitory computer-readable memory storing further instructions executed by the at least one processor (Karri, para. [0095]-[0096])
System claim 19 and method claim 11 are related as system and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 19 is similarly rejected under the same rationale as applied above with respect to claim 11.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO 892 form.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to OLUJIMI A ADESANYA whose telephone number is (571)270-3307. The examiner can normally be reached Monday-Friday 8:30-5:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Richemond Dorvil can be reached at 571-272-7602. 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.
/OLUJIMI A ADESANYA/Primary Examiner, Art Unit 2658
Prosecution Insights