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 .
Priority
Acknowledgment is made of applicant's claim for foreign priority based on an application filed in Romania on 09/26/2024. It is noted, however, that applicant has not filed a certified copy of the ROA 2024 00576 application as required by 37 CFR 1.55.
Response to Arguments
Applicant's arguments filed 08/24/2026 have been fully considered but they are not persuasive. Regarding arguments on pages 1-2 of the Remarks, Examiner notes that the changing of the information supplied to the second service could be interpreted as a mental process. A human could use the components and the received values to modify the components. The claims do not detail how the modification is performed, leaving the limitation open to a broad interpretation. Further, the claims lack detail on the action to be performed, merely that an action is triggered by sending an executable command. Without details about the action, the limitations include mental processes. For example, a human could perform an action upon receiving a command. Therefore, the claim limitations are still characterized as abstract, and do not include sufficient detail to be considered implemented into a practical application or significantly more.
Regarding arguments on pages 3-4 of the Remarks, Examiner notes that the claims do not require that the first and second natural language components be supplied to different processing services. However, in Assa, the speech understanding model is interpreted to provide services including intent generation and slot generation. Assa Fig. 6A shows the configurable entities, their objects, and corresponding property values. Upon receiving a speech input, the system determines the intents and slots. Para [0129], [0132] and Fig. 7A of Assa show examples of how the intent includes the change to be made, while the slots include the details on how the change is to be made. For example, a change_glasses intent is determined as the first response value, and based on the determined intent or value, the slots are determined and filled, including determining the objects or slots of “glasses_color” and “glasses_material”, and the property values from the second part of the input of “red” and “plastic”. Therefore the ordered relationship is satisfied, as the property values and objects depend on the configurable entity that is first determined.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-6 and 8-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. Using the subject matter eligibility test from page 74621 of the Federal Register Notice titled “2014 Interim Guidance on Patent Subject Matter Eligibility,” a two-step process is performed. Under step 1, the claims are analyzed to determine if the claim is directed to a process, machine, article of manufacture, or composition of matter. In this case, claims 1-10 are directed to a method, which is a process; claims 11-15 are directed to a computer-program product, which is a machine or an article of manufacture; claims 16-20 are directed to system, which is a machine or an article of manufacture. Step 2A (part 1 of the Mayo test), using the guidance from pages 50-57 of the Federal Register Vol. 84 No. 4 from Monday, January 7, 2019, requires applying a two-prong inquiry. In Prong One, examiners evaluate whether the claim recites a judicial exception, determining if the claim is directed to a law of nature, a natural phenomenon, or an abstract idea. In this case, claim 1 recites receiving a request, determining a logical dependency, determining categories, selecting a service, generating a modified natural language component, and triggering an action, which is are mental processes. In Prong Two, examiners evaluate whether the judicial exception is integrated into a practical application that imposes a meaningful limit on the judicial exception. In this case, limitations of sending and receiving data are mere extrasolution activity, and do not integrate the abstract ideas into a practical application.
Step 2B (part 2 of the Mayo test) requires analyzing the claims to determine if they recite additional elements that amount to significantly more than the judicial exception. In this case, the claims do not include additional elements that are sufficient to amount to significantly more than the abstract idea itself.
Regarding claims 1, 11, and 16, receiving a request, determining a logical dependency, determining categories, selecting a service, generating a modified natural language component, and triggering an action are mental processes, which is an abstract idea. For example, a human could receive a request in natural language, could determine a logical dependency in the request, determine categories for the natural language components, select which service to use to process the components, could use the results to generate a modified natural language component, and could trigger an action to be performed. Additional limitations of sending and receiving data are mere extrasolution activity, while structural elements of processor and computer-readable media are generic computing components, and do not integrate the abstract ideas into a practical application or constitute significantly more.
Regarding claims 2, 12, and 17, performing confirmation is a mental process, which is an abstract idea, while displaying information is mere extrasolution activity, and does not integrate the abstract idea into a practical application or constitute significantly more.
Regarding claims 3, 13, and 18, accessing metadata and generating a prompt are mental processes, which is an abstract idea. Prompting a large language model is mere extrasolution activity, as this merely involves transmitting the data, and does not integrate the abstract ideas into a practical application or constitute significantly more.
Regarding claims 4, 14, and 19, prompting a large language model and receiving a response are mere extrasolution activity, and do not integrate the abstract ideas into a practical application or constitute significantly more.
Regarding claims 5, 15, and 20, accessing rules and determining separation of natural language components are mental processes, which is an abstract idea, without integration into a practical application and without significantly more. For example, a human could parse a sentence and separate sections using rules.
Regarding claims 6 and 8, generating a prompt and determining a category or action are mental processes, which is an abstract idea. Prompting a large language model is mere extrasolution activity, as this merely involves transmitting the data, and does not integrate the abstract ideas into a practical application or constitute significantly more.
Regarding claim 9, displaying, sending, and receiving data are mere extrasolution activity, and do not integrate the abstract ideas into a practical application or constitute significantly more.
Regarding claim 10, causing navigation to a particular interface is mere extrasolution activity, and do not integrate the abstract ideas into a practical application or constitute significantly more.
The limitations of the claims, taken alone, do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements individually. Applicable case law cited in the Federal Register includes, but is not limited to: Alice Corp., 134 S. Ct. at 2355-56, Digitech Image Tech., LLC v. Electronics for Imaging, Inc., 758 F.3d 1344 (Fed. Cir. 2014), Benson, 409 U.S. at 63.
See "Preliminary Examination Instructions in view of the Supreme Court Decision in Alice Corporation Pty. Ltd. v. CLS Bank International, et al.," dated June 25, 2014, and the Federal Register notice titled "2014 Interim Guidance on Patent Subject Matter Eligibility" (79 FR 74618).
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1, 5, 7, 9-11, 15-16, and 20 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Assa et al. (US 2023/0067305 A1), hereinafter referred to as Assa.
Regarding claim 1, Assa teaches:
A computer-implemented method comprising:
receiving a request comprising a natural language item, the natural language item comprising a first natural language component and a second natural language component (Fig. 7A, para [0129], [0132], where a natural language request is received, and a set of intents and slots are generated);
determining a logical dependency of the second natural language component on the first natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing);
determining a first category for the first natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the intent include putting or removing mascara or lipstick);
determining a second category for the second natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the slots include color, material, type, and width);
selecting, from a plurality of different candidate natural language processing services, a first natural language processing service associated with the first category, wherein the first natural language processing service supports one or more first applications (para [0039], where the speech understanding model processes the speech and generates estimated intents, the first service being the intent generation, and para [0133], where a makeup application is used);
selecting, from a plurality of different candidate natural language processing services, a second natural language processing service associated with the second category, wherein the second natural language processing service supports one or more second applications (para [0039], where the speech understanding model processes the speech and generates estimated slots, the second service being the slot generation, and para [0133], where a makeup application is used);
sending the first natural language component to the first natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated intents);
receiving a first response from the first natural language processing service, wherein the first response comprises one or more values responsive to the first natural language component (Fig. 7A, para [0129], 0132], where the first response is the generated intent);
based at least in part on the second natural language component and the one or more values responsive to the first natural language component, generating a modified second natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing, the modified component including the determination of the object and the original second natural language component);
sending the modified second natural language component to the second natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated slots);
receiving a second response from the second natural language processing service, wherein the second response comprises an option to trigger one or more actions in at least one second application of the one or more second applications responsive to the modified second natural language component (Fig. 7A, para [0129], 0132], where the second response is the generated update to the configurable entity, which is used to change an object in a graphical user interface such as from makeup application of para [0133]);
triggering the one or more actions by sending one or more executable commands for performing the one or more actions to the at least one second application (para [0133], where the makeup application applies the update).
Regarding claim 5, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
accessing one or more stored rules to detect separation between natural language components (para [0120-122], where the machine learning uses rules to generate updates for intent/slot from the raw speech input);
determining the first natural language component and the second natural language component from the natural language item based at least in part on a particular detected separation between the first natural language component and the second natural language component (para [0125], where the speech understanding model generates the estimated intents and slots, such as the intent being to change glasses and the slot being the color blue).
Regarding claim 7, Assa teaches:
The computer-implemented method of Claim 1, wherein the first natural language processing service comprises a first machine learning model (para [0110], where the speech understanding model is trained on speech inputs and learns to generate intents and entities for determining intents),
wherein the first machine learning model is trained based at least in part on first metadata that describes one or more first candidate operations for the one or more first applications (para [0038], [0111], where the machine learning model is trained based on the schemas, which identify the states of the entities and their property values),
wherein the second natural language processing service comprises a second machine learning model (para [0110], where the speech understanding model is trained on speech inputs and learns to generate intents and entities for determining slots), and
wherein the second machine learning model is trained based at least in part on second metadata that describes one or more second candidate operations for the one or more second applications (para [0038], [0111], where the machine learning model is trained based on the schemas, which identify the states of the entities and their property values).
Regarding claim 9, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
causing display of the one or more actions to a user (Fig. 7B, para [0130], where an action results in displaying glasses with transparent lens);
receiving an input from the user indicating against the one or more actions (Fig. 7B, para [0131], where the user requests darker frames on the glasses, going against the displayed glasses); and
to reverse the one or more actions, sending to the at least one second application another one or more commands associated with the one or more commands (Fig. 7B, para [0131], where the display is updated to include the change or reversal to the result of the first action).
Regarding claim 10, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
causing navigation to a particular interface in the at least one second application, wherein the one or more actions partially accomplish the request defined by the natural language item, and wherein the particular interface comprises one or more options to complete the request defined by the natural language item (Fig. 7A, para [0129], [0132], where the application GUI displays the augmented reality, with the change from the request reflected in the updated GUI, the options including the possible alterations that can be made).
Regarding claim 11, Assa teaches:
A computer-program product comprising one or more non-transitory machine-readable storage media (Fig. 9 element 916, para [0144], where instructions are stored on a machine-readable medium), including stored instructions configured to cause a computing system to perform a set of actions including:
receiving a request comprising a natural language item, the natural language item comprising a first natural language component and a second natural language component (Fig. 7A, para [0129], [0132], where a natural language request is received, and a set of intents and slots are generated);
determining a logical dependency of the second natural language component on the first natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing);
determining a first category for the first natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the intent include putting or removing mascara or lipstick);
determining a second category for the second natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the slots include color, material, type, and width);
selecting, from a plurality of different candidate natural language processing services, a first natural language processing service associated with the first category, wherein the first natural language processing service supports one or more first applications (para [0039], where the speech understanding model processes the speech and generates estimated intents, the first service being the intent generation, and para [0133], where a makeup application is used);
selecting, from a plurality of different candidate natural language processing services, a second natural language processing service associated with the second category, wherein the second natural language processing service supports one or more second applications (para [0039], where the speech understanding model processes the speech and generates estimated slots, the second service being the slot generation, and para [0133], where a makeup application is used);
sending the first natural language component to the first natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated intents);
receiving a first response from the first natural language processing service, wherein the first response comprises one or more values responsive to the first natural language component (Fig. 7A, para [0129], 0132], where the first response is the generated intent);
based at least in part on the second natural language component and the one or more values responsive to the first natural language component, generating a modified second natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing, the modified component including the determination of the object and the original second natural language component);
sending the modified second natural language component to the second natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated slots);
receiving a second response from the second natural language processing service, wherein the second response comprises an option to trigger one or more actions in at least one second application of the one or more second applications responsive to the modified second natural language component (Fig. 7A, para [0129], 0132], where the second response is the generated update to the configurable entity, which is used to change an object in a graphical user interface such as from makeup application of para [0133]);
triggering the one or more actions by sending one or more executable commands for performing the one or more actions to the at least one second application (para [0133], where the makeup application applies the update).
Regarding claim 15, Assa teaches:
The computer-program product of Claim 11, wherein the set of actions further includes:
accessing one or more stored rules to detect separation between natural language components (para [0120-122], where the machine learning uses rules to generate updates for intent/slot from the raw speech input);
determining the first natural language component and the second natural language component from the natural language item based at least in part on a particular detected separation between the first natural language component and the second natural language component (para [0125], where the speech understanding model generates the estimated intents and slots, such as the intent being to change glasses and the slot being the color blue).
Regarding claim 16, Assa teaches:
A system comprising:
one or more processors (Fig. 9 element 902, para [0143], where processors are used);
one or more non-transitory computer-readable media storing instructions (Fig. 9 element 916, para [0144], where a machine-readable medium is used), which, when executed by the system, cause the system to perform a set of actions including:
receiving a request comprising a natural language item, the natural language item comprising a first natural language component and a second natural language component (Fig. 7A, para [0129], [0132], where a natural language request is received, and a set of intents and slots are generated);
determining a logical dependency of the second natural language component on the first natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing);
determining a first category for the first natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the intent include putting or removing mascara or lipstick);
determining a second category for the second natural language component (Fig. 7A, para [0108], [0129], 0132], where categories for the slots include color, material, type, and width);
selecting, from a plurality of different candidate natural language processing services, a first natural language processing service associated with the first category, wherein the first natural language processing service supports one or more first applications (para [0039], where the speech understanding model processes the speech and generates estimated intents, the first service being the intent generation, and para [0133], where a makeup application is used);
selecting, from a plurality of different candidate natural language processing services, a second natural language processing service associated with the second category, wherein the second natural language processing service supports one or more second applications (para [0039], where the speech understanding model processes the speech and generates estimated slots, the second service being the slot generation, and para [0133], where a makeup application is used);
sending the first natural language component to the first natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated intents);
receiving a first response from the first natural language processing service, wherein the first response comprises one or more values responsive to the first natural language component (Fig. 7A, para [0129], 0132], where the first response is the generated intent);
based at least in part on the second natural language component and the one or more values responsive to the first natural language component, generating a modified second natural language component (Fig. 6A, para [0107-108], where each property value depends on the object it is describing, the modified component including the determination of the object and the original second natural language component);
sending the modified second natural language component to the second natural language processing service (para [0039], where the speech understanding model processes the speech and generates estimated slots);
receiving a second response from the second natural language processing service, wherein the second response comprises an option to trigger one or more actions in at least one second application of the one or more second applications responsive to the modified second natural language component (Fig. 7A, para [0129], 0132], where the second response is the generated update to the configurable entity, which is used to change an object in a graphical user interface such as from makeup application of para [0133]);
triggering the one or more actions by sending one or more executable commands for performing the one or more actions to the at least one second application (para [0133], where the makeup application applies the update).
Regarding claim 20, Assa teaches:
The system of Claim 16, wherein the set of actions further includes:
accessing one or more stored rules to detect separation between natural language components (para [0120-122], where the machine learning uses rules to generate updates for intent/slot from the raw speech input);
determining the first natural language component and the second natural language component from the natural language item based at least in part on a particular detected separation between the first natural language component and the second natural language component (para [0125], where the speech understanding model generates the estimated intents and slots, such as the intent being to change glasses and the slot being the color blue).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 2, 12, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Assa, in view of Schweinfurth (US 11,430,051 B1).
Regarding claim 2, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
Assa does not teach:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user.
Schweinfurth teaches:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user (Figs. 15-16, col. 22 lines 31-41, col. 23 lines 1-48, where a confirmation window is displayed, and where the action is performed upon receiving user confirmation).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the confirmation of Schweinfurth (Schweinfurth Fig. 15) for the actions of Assa (Assa para [0133]), in order to provide the user with the information to confirm before agreeing to trigger the action (Schweinfurth col. 23 lines 1-33).
Regarding claim 12, Assa teaches:
The computer-program product of Claim 11, wherein the set of actions further includes:
Assa does not teach:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user.
Schweinfurth teaches:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user (Figs. 15-16, col. 22 lines 31-41, col. 23 lines 1-48, where a confirmation window is displayed, and where the action is performed upon receiving user confirmation).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the confirmation of Schweinfurth (Schweinfurth Fig. 15) for the actions of Assa (Assa para [0133]), in order to provide the user with the information to confirm before agreeing to trigger the action (Schweinfurth col. 23 lines 1-33).
Regarding claim 17, Assa teaches:
The system of Claim 16, wherein the set of actions further includes:
Assa does not teach:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user.
Schweinfurth teaches:
causing the option to trigger the one or more actions in the at least one second application to be displayed to a user, and wherein the triggering the one or more actions is in response to receiving a confirmation from the user (Figs. 15-16, col. 22 lines 31-41, col. 23 lines 1-48, where a confirmation window is displayed, and where the action is performed upon receiving user confirmation).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the confirmation of Schweinfurth (Schweinfurth Fig. 15) for the actions of Assa (Assa para [0133]), in order to provide the user with the information to confirm before agreeing to trigger the action (Schweinfurth col. 23 lines 1-33).
Claim(s) 3-4, 6, 13-14, and 18-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Assa, in view of Manandise et al. (US 2025/0139367 A1), hereinafter referred to as Manandise.
Regarding claim 3, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
accessing metadata that describes one or more candidate operations for the one or more first applications (para [0038], where a schema is obtained which identifies a current state of the entity and a list of expected intents and slots);
Assa does not teach:
generating a prompt including at least part of the metadata and the first response;
prompting a large language model with the prompt to validate that the first response is a valid response for the first category.
Manandise teaches:
generating a prompt including at least part of the metadata and the first response (para [0050-52], where the extracted information is interpreted as the first response, while the domain description is interpreted as the metadata, the task description in PDDL being the prompt);
prompting a large language model with the prompt to validate that the first response is a valid response for the first category (para [0053], where a validation component determines whether the output conforms to the layout and/or elements expected, such as by comparison to a template).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the validation of Manandise (Manandise para [0053]) on the request of Assa (Assa para [0129], [0132]), in order to detect errors and provide feedback to generate new prompts (Manandise para [0053]).
Regarding claim 4, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
Assa does not teach:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item; and
receiving the first natural language component and the second natural language component from the large language model.
Manandise teaches:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item (para [0044], where a prompt is parsed to analyze the grammatical structure, and the head and dependent are each labeled); and
receiving the first natural language component and the second natural language component from the large language model (para [0047], where the LLM generates task descriptions in PDDL using the categorization of the tokens of the prompt).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the parsing of Manandise (Manandise para [0044]) on the request of Assa (Assa para [0129], [0132]), in order to determine relationships in the text for understanding the meaning of the text (Manandise para [0044]).
Regarding claim 6, Assa teaches:
The computer-implemented method of Claim 1, wherein the determining the first category for the first natural language component comprises:
generating a prompt, wherein the prompt comprises the first natural language component and a plurality of candidate categories (para [0038], where the raw speech input and the schema are provided to the speech understanding model, the schema including the plurality of candidate categories); and
Assa does not teach:
prompting a large language model with the prompt to determine a particular category of the plurality of candidate categories for the first natural language component; and
based at least in part on a result of the prompting, determining the first category for the first natural language component.
Manandise teaches:
prompting a large language model with the prompt to determine a particular category of the plurality of candidate categories for the first natural language component (para [0046-47], where a prompt is tokenized, and the LLM categorizes the tokens); and
based at least in part on a result of the prompting, determining the first category for the first natural language component (para [0046-47], where a prompt is tokenized, and the LLM categorizes the tokens).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the parsing of Manandise (Manandise para [0044]) on the request of Assa (Assa para [0129], [0132]), in order to determine relationships in the text for understanding the meaning of the text (Manandise para [0044]).
Regarding claim 13, Assa teaches:
The computer-program product of Claim 11, wherein the set of actions further includes:
accessing metadata that describes one or more candidate operations for the one or more first applications (para [0038], where a schema is obtained which identifies a current state of the entity and a list of expected intents and slots);
Assa does not teach:
generating a prompt including at least part of the metadata and the first response;
prompting a large language model with the prompt to validate that the first response is a valid response for the first category.
Manandise teaches:
generating a prompt including at least part of the metadata and the first response (para [0050-52], where the extracted information is interpreted as the first response, while the domain description is interpreted as the metadata, the task description in PDDL being the prompt);
prompting a large language model with the prompt to validate that the first response is a valid response for the first category (para [0053], where a validation component determines whether the output conforms to the layout and/or elements expected, such as by comparison to a template).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the validation of Manandise (Manandise para [0053]) on the request of Assa (Assa para [0129], [0132]), in order to detect errors and provide feedback to generate new prompts (Manandise para [0053]).
Regarding claim 14, Assa teaches:
The computer-program product of Claim 11, wherein the set of actions further includes:
Assa does not teach:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item; and
receiving the first natural language component and the second natural language component from the large language model.
Manandise teaches:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item (para [0044], where a prompt is parsed to analyze the grammatical structure, and the head and dependent are each labeled); and
receiving the first natural language component and the second natural language component from the large language model (para [0047], where the LLM generates task descriptions in PDDL using the categorization of the tokens of the prompt).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the parsing of Manandise (Manandise para [0044]) on the request of Assa (Assa para [0129], [0132]), in order to determine relationships in the text for understanding the meaning of the text (Manandise para [0044]).
Regarding claim 18, Assa teaches:
The system of Claim 16, wherein the set of actions further includes:
accessing metadata that describes one or more candidate operations for the one or more first applications (para [0038], where a schema is obtained which identifies a current state of the entity and a list of expected intents and slots);
Assa does not teach:
generating a prompt including at least part of the metadata and the first response;
prompting a large language model with the prompt to validate that the first response is a valid response for the first category.
Manandise teaches:
generating a prompt including at least part of the metadata and the first response (para [0050-52], where the extracted information is interpreted as the first response, while the domain description is interpreted as the metadata, the task description in PDDL being the prompt);
prompting a large language model with the prompt to validate that the first response is a valid response for the first category (para [0053], where a validation component determines whether the output conforms to the layout and/or elements expected, such as by comparison to a template).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the validation of Manandise (Manandise para [0053]) on the request of Assa (Assa para [0129], [0132]), in order to detect errors and provide feedback to generate new prompts (Manandise para [0053]).
Regarding claim 19, Assa teaches:
The system of Claim 16, wherein the set of actions further includes:
Assa does not teach:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item; and
receiving the first natural language component and the second natural language component from the large language model.
Manandise teaches:
prompting a large language model with the natural language item to detect the first natural language component and the second natural language component from the natural language item (para [0044], where a prompt is parsed to analyze the grammatical structure, and the head and dependent are each labeled); and
receiving the first natural language component and the second natural language component from the large language model (para [0047], where the LLM generates task descriptions in PDDL using the categorization of the tokens of the prompt).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the parsing of Manandise (Manandise para [0044]) on the request of Assa (Assa para [0129], [0132]), in order to determine relationships in the text for understanding the meaning of the text (Manandise para [0044]).
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Assa, in view of Wang et al. (US 12,433,527 B1), hereinafter referred to as Wang.
Regarding claim 8, Assa teaches:
The computer-implemented method of Claim 1, wherein the method further comprises:
Assa does not teach:
generating a prompt comprising the one or more commands;
prompting a large language model to generate a natural language description of the one or more commands; and
determining the one or more actions based at least in part on a response from the prompting.
Wang teaches:
generating a prompt comprising the one or more commands (col. 12 lines 52-67, where a prompt comprises predicted treatment responses);
prompting a large language model to generate a natural language description of the one or more commands (col. 12 lines 52-67, where the LLM generates natural language descriptions of the treatment recommendation); and
determining the one or more actions based at least in part on a response from the prompting (col. 12 lines 52-67, where the treatment recommendation is displayed).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Assa by using the LLM of Wang (Wang col. 12 lines 52-67) to explain the commands of Assa (Assa para [0133]), in order to enable clinicians to identify effective treatments for individual patients or choose an optimal treatment based on predicted response (Wang col. 2 lines 18-48).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. US 10,600,406 B1 col. 6 line 63 – col. 7 line 15, where each intent specifies a list of slots.
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 BRYAN S BLANKENAGEL whose telephone number is (571)270-0685. The examiner can normally be reached 8:00am-5:30pm.
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.
/BRYAN S BLANKENAGEL/Primary Examiner, Art Unit 2658
Prosecution Insights