DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This application has been examined. Claims 1-19 are pending.
The prior art submitted on 4/21/25 and 12/1/25 has been considered.
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.
Claims 1-2,4, 9-10,12, and 16-19, are rejected under 35 U.S.C. 103 as being unpatentable over Gurgu et al. (US 2023/0297887 A1) in view of Peng et al. (US 2024/0070394 A1).
As per claims 1, and 19, Gurgu et al. disclose method, and a non-transitory computer readable medium, for reducing processing complexity for command and control of a remote autonomous object in response to natural language commands (see at least para. [0031] disclose more specifically, for in-context learning of a given task, the language model may receive as an input prompt, a description of the task, an optional number of labeled examples demonstrating the task), the method comprising: storing, in memory, a prompt library, a conversational library, program code for generating one or more neural networks for producing a complex command message (see at least [0066] disclose the inference models used by the intent classification system 100 may include, for example, deep neural network, shallow neural networks, and the like. The neural networks may have an input layer, one or more hidden layers, and an output layer. One or more of the neural networks may generate one or more embeddings (also referred to as features) from the user query);
receiving, at a basic processor, a user command in a natural language format (see at least [0065] disclose the intent classification system 100 may include one or more machine learning models (also referred to as inference models) that are trained to identify a user intent based on user query. For example, the intent classification system 100 may receive a query such as “What is my order status,” “I need to make a payment,” or “Can I get a refund on my item,” and output a predicted intent for the query, such as, for example, “order status,” “make payment,” or “get refund.” A response or answer may be output based on the classified item.”);
extracting, by the basic processor, designated parameters from the user command for matching with a control prompt stored in a library of control prompts (see at least [0061] disclose the end user may formulate a query, and transmit the query to the chatbot system 10 as a chat message, text message, social media message, and/or the like. The chatbot system 10 may process the query and determine a user intent. One or more machine learning models may be invoked for predicting the user intent. Once the intent is determined, the chatbot may output an answer in response to the query. Th one or more machine learning models, and software and hardware are for interfacing with the end user devices 16, may generally be referred to as a chatbot; also para. [0091] disclose in on embodiment, the characteristic determined by the filtering system 302 is semantic and/or lexical similarity of the generated question to the input prompt. In one embodiment, filtering system 302 generates n-grams of the words contained in at least a portion of the prompt (e.g., answer title and/or answer content), and n-grams of the words contained in the generated question. The filtering system 302 may compare the n-grams to determine overlap between the generated question and the prompt. The amount of overlap in the n-grams may be used as an indication of semantic relevance of the generated question to the prompt containing at least a portion of the answer. In addition or in lieu of n-grams, a cosine similarly measure may be used to compute the semantic similarity between the generate question and the prompt; and para. [0097] disclose in some embodiments, a plurality of answer content is sampled for generating a plurality of prompts. The prompts are provided to the large language model for receiving various question suggestions on the prompts), and determining a confidence factor with which the matching has been performed relative to a predetermined threshold (see at least [0094] disclose in one embodiment, the filtering system 302 discards the generated question in response to the lexical and/or semantic relevance being below a threshold; and para. [0119] disclose in some embodiments, one or more tokens (e.g., words of the suggested question) are selected based on the in-context learning. In this regard, a probability score may be computed for each token within a vocabulary of the model based on training received by the model, and one or more tokens may be selected based on the probability score);
selectively routing, by the basic processor, the user command to a simple command processing path or a complex command processing path based on the confidence factor, the simple command processing path configured for generating a simple command message that includes a simple command objective generated from the control prompt matched to the designated parameters and a conversational counterpart retrieved from a conversational library (see at least [0061] disclose for example, the end user may formulate a query, and transmit the query to the chatbot system 10 as a chat message, text message, social media message, and/or the like. The chatbot system 10 may process the query and determine a user intent. One or more machine learning models may be invoked for predicting the user intent. Once the intent is determined, the chatbot may output an answer in response to the query. The one or more machine learning models, and software and hardware for interfacing with the end user devices 16, may generally be referred to as a chatbot; also para. [0097] disclose the prompts are provided to the large language model for receiving various question suggestions based on the prompts. In some embodiments, the evaluation system 304 computes one or more metrics for the generated question suggestions. A parameter of the large language model may be altered based on the one or more metrics. For example, a hyperparameter of the language model may be altered. In some embodiments, the alteration may be of the prompt structure. In yet some embodiment, the current large language model may be replaced with a different large language model based on the computed metrics),
and the complex command processing path having one or more neural networks that process the user command to extract library content identifying context and situational awareness that is semantically related to the user command, and generate a complex command message based on the library content; comparing, by the basic processor, the simple command objective message or the complex command objective message generated because of the selective routing (see at least para. [0055] disclose in this regard, the prompt may include a topic of discussion typical for the enterprise, text describing a context of the discussion, and describe the task as generating a question about the topic of discussion. The language model may engage in zero-shot or few-shot in-context learning to fulfill the described task; and para. [0061] disclose the end user device 16 may be a desktop, laptop, and/or any other computing device conventional in the art. A customer, potential customer, or other end user (collectively referenced as an end user) desiring to receive services from the enterprise may initiate communications to the chatbot system 10 using the end user device 16; and para. [0060] disclose the chatbot system 10 is configured to handle interactions with the end user device 16. The chatbot system 10 may be configured to handle interactions on behalf of a particular business or enterprise, or on behalf of multiple businesses or enterprises. For example, a separate instance of a chatbot system 10 may be provided for each separate enterprise for handling interactions of the enterprise). Gurgu et al. do not explicitly disclose known capabilities of the remote autonomous object to identify a remote autonomous object command; and generating a remote autonomous object command message containing the remote autonomous object command for transmission to the remote autonomous object. However, Peng et al. disclose known capabilities of the remote autonomous object to identify a remote autonomous object command; and generating a remote autonomous object command message containing the remote autonomous object command for transmission to the remote autonomous object (see at least para. [0062] disclose for example, in one embodiment, ser device 510 may be implemented as an autonomous driving vehicle; para. [0064] disclose for example, other applications 516 may contain software programs for asset management executable by a processor, including a graphical user interface (GUI) configured to provide an interface to the user 540 to view the output, such as an answer to an input question; and para. [0068] disclose for example, in one implementation, the data vendor server 545 may send asset information from the database 519, via the network interface 526, to the server 530). It 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 to modify the teach of Gurgu et al. by combining known capabilities of the remote autonomous object to identify a remote autonomous object command; and generating a remote autonomous object command message containing the remote autonomous object command for transmission to the remote autonomous object to provide a mechanism that ensembles soft prompts to transfer knowledge from source tasks under few-shot learning setting, in order to optimize performance of executable actions based on the applicable context.
As per claim 2, Gurgu et al. disclose selectively routing the user command to the simple command processing path when the confidence factor meets or exceeds the predetermined threshold (see at least [0101] disclose in one embodiment, in computing the similarity to the question suggestion to the answer content, the evaluation system 304 computes a cosine similarity distance between the embeddings generated for the question suggestion, and the embeddings generated for the answer content. The question suggestion may be accepted as being semantically relevant if the computed cosine similarity distance is below a threshold. In some embodiments, the evaluation system 304 assigns an appropriate label to the question suggestion as, for example, rejected or accepted, based on the semantic relevance determination; also para. [0119] disclose in act 406, the input is provided to the machine learning model. In this regard, the machine learning model may engage in in-context learning where the machine learning model learns how to perform a task by conditioning on the input. In some embodiments, one or more tokens (e.g., words of the suggested question) are selected based on the in-context learning. In this regard, a probability score may be computed for each token within a vocabulary of the model based on training received by the model, and one or more tokens may be selected based on the probability score).
As per claim 4, Gurgu et al. disclose selectively routing the user command to the complex command processing path when the confidence factor is below the predetermined threshold (see at least para. [0052] disclose it should be appreciated that the performance of the language model for accurately making predictions that fulfill a particular task may depend on the types of prompts provided to the language model, and/or the labeled examples included in the prompts. For example, by selecting the appropriate prompts, an administrator may be able to manipulate a model’s behavior so that the language model may be used to predict the desired output even without any task-specific training; also para. [0053] disclose in some embodiment, the language model used for in-context learning is a generative language model. The task given to the generative language model may be to generate synthetic training data using the input prompt. The use of a generative language model in a traditional manner, however, may not be without limitation; and para. [0122] disclose in some embodiments, the predicted characteristic includes textual similarity of the output to all or a portion of the input (e.g., the answer title and/or answer content). In this regard, the filtering system 302 may compute semantic similarity and/or lexical similarity scores to determine whether the output is textually similar to at least a portion of the input. The output may be filtered in act 412 in response to the similarity measure being below a threshold value).
As per claim 16, Gurgu et al. do not explicitly disclose the remote autonomous vehicle. However, Peng et al. disclose mounted on the remote autonomous vehicle (see at least [0062] disclose for example, in one embodiment, user device 510 may be implemented as an autonomous driving vehicle). It 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 to modify the teach of Gurgu et al. by combining the remote autonomous vehicle to process task in order to optimize performance of executable actions based on the applicable context.
As per claim 17, Gurgu et al. do not explicitly disclose mounted in a human-wearable article. However, Peng et al. disclose mounted in a human-wearable article (see at least para. [0062] disclose for example, in one embodiment, user device 510 may be implemented as an autonomous driving vehicle, a personal computer (PC), a smart phone, laptop/tablet computer, wristwatch with appropriate computer hardware resources, eyeglasses with appropriate computer hardware (e.g., GOOGLE GLASSES), other type of wearable computing device). It 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 to modify the teach of Gurgu et al. by combining human-wearable article to process task in order to optimize performance of executable actions based on the applicable context.
As per claim 18, Gurgu et al. disclose mounted in a server configured to receive the user command over a network (see at least [0062] disclose in one embodiment, the chatbot builder 12 may include a computing system that is used by a chatbot administrator to configure, train, and/or maintain, for a particular enterprise, one or more machine learning model (also referred to as inference models) of the chatbot system 10. The computing system may be a desktop computer, laptop computer, network server, mobile device, embedded computer, and/or the like).
Claims 9-10, and 12, are system claims corresponding to method claims 1-2, and 4 above. Therefore, they are rejected for the same rationales set forth as above.
Claims 3, 5-8, 11, and 13-15, are rejected under 35 U.S.C. 103 as being unpatentable over Gurgu et al. (US 2023/0297887 A1), and Peng et al. (US 2024/0070394 A1), as applied to claim 1 above, and further in view of Gloginic et al. (US 2025/0292079 A1).
As per claim 3, Gurgu et al. disclose wherein generating the simple command message comprises: generating the simple command objective from the control prompt matched to the extracted designated parameters; identifying counterpart command terms in the conversational library using the extracted designated parameters (see at least para. [0053] disclose for example, because a generative language model is a black box solution, there may be a general lack of interpretability of the outputs that is produces. In addition, the general difficulty to control the generation process may require careful prompt engineering and experimentation; also para. [0128] disclose in some embodiment, the parameters to be tested may relate to the prompt structure. For example, certain test wording may be used for the prompt structure to determine its effectiveness in generating relevant training questions; and para. [0133] disclose in act 612, the evaluation system 304 may be done testing when a set of different values identified for evaluation have been evaluated. If the answer is YES, the evaluation system 304 selects the value for the parameter resulting in the highest metrics (e.g., highest aggregate of the UGR, MSAV, and MSAR metrics), as the final value for the corresponding parameter of the large language model). Gurgu et al. do not explicitly disclose combining the simple command objective and the counterpart command terms into a processed simple command message. However, Gloginic et al. disclose combining the simple command objective and the counterpart command terms into a processed simple command message (see at least the abstract; and para. [0084-0087] disclose the task may include one or more response, generated by the LLM/MLM, to the one or more LLM prompts (or other MLM inputs) of the task data (e.g., one or more inference responses to prompts received by a customer-facing chatbot). It 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 to modify the teach of Gurgu et al. by combining combine the simple command objective and the counterpart command terms into a processed simple command message in order to optimize performance of executable actions based on the applicable context.
As per claim 5, Gurgu et al. disclose generating the complex command message comprises: executing, by the basic processor, the program code stored in memory, the program code causing the basic processor to generate at least: a large language model (LLM) for producing an LLM prompt (see at least para. [0097] disclose in some embodiments, a plurality of answer content is sampled for generating a plurality of prompts. The prompts are provided to the large language model for receiving various question suggestions based on the prompts. In some embodiment, the evaluation system 304 computes one or more metrics for the generated question suggestions. A parameter of the large language model may be altered based on the one or more metrics), and the LLM being further configured for generating the complex command objective; passing the user command to the LLM to build the LLM prompt, the LLM querying the RAG model for context data that is semantically related to the user command (see at least para. [0110] disclose in some embodiments, an aggregate of the one or more metrics computed for a first large language model is compared against an aggregate of the one or more metrics computed for a second large language model. The first large language model may be different from the second large language model in terms of type, and/or value of one or more hyperparameters. The prompt structure used to generate a prompt for the first large language model may be the same or different from the prompts structure used to generate a prompt for the second large language model);
returning, by the RAG model, the context data package including semantically related context data to the LLM (see at least para. [0122] disclose in some embodiments, the predicted characteristic includes textual similarity of the output to all or a portion of the input (e.g., the answer title and/or answer content). In this regard, the filtering system 302 may compute semantic similarity and/or lexical similarity scores to determine whether the output is textually similar to at least a portion of the input. The output may be filtered in act 412 in response to the similarity measure being below a threshold value); and passing the LLM prompt to an input of the LLM model to generate a complex command objective message (see at least para. [0130] disclose in act 606, the large language model to generated a suggestion question based on the input prompt. In some embodiment, multiple suggested training questions are generated (e.g., serially) based on the input prompt). Gurgu et al. do not explicitly disclose a retrieval augmented generation (RAG) model for producing a context data package, and combining, by the LLM, the user command and the data package to generate the LLM prompt. However, Gloginic et al. disclose a retrieval augmented generation (RAG) model for producing a context data package, and combining, by the LLM, the user command and the data package to generate the LLM prompt (see at least para. [0020], and [0169-0173], all para. disclose a RAG component 1192 (which may include one or more RAG models, and/or may be performed using the generative LM 1130 itself) may be used to retrieve additional information to be used as part of the input 1101 or prompt). It 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 to modify the teach of Gurgu et al. by combining a retrieval augmented generation (RAG) model for producing a context data package, and combining, by the LLM, the user command and the data package to generate the LLM prompt in order to optimize performance of executable actions based on the applicable context.
As per claim 6, Gurgu et al. disclose comparing, by the RAG model, the user command with context data stored in memory, and identifying one or more context data elements that are semantically related to the user command and that specify a related context and situational awareness of the user command (see at least para. [0089] disclose in some embodiment, the recommendation system 300 selects a prompt structure from one or more different prompt structures based on a predicted success in generating a training question. The success may be based on a semantic similarity measure. For example, one prompt structure may use the wording “The item to be discussed is” while another prompt structure may be use the wording “The topic of discussion is”. The prompt structure that provide a training question with a higher semantic similarity measure to the input prompt may be selected for use by the recommendation system 300; also para. [0110] disclose in some embodiment, an aggregate of the one or more metrics computed for a first large language model is compared against an aggregate of the one or more metrics computed for a second large language model. The first large language model may be different from the second large language model in terms of type, and/or value of one or more hyperparameters. The prompt structure used to generate a prompt for the first large language model may be the same or different from the prompt structure used to generated a prompt for the second large language model).
As per claim 7, Peng et al. disclose the context data stored in memory comprises: a Pre-Mission Upload including Constitution and Warfighter Operational Context, ethical guidelines, a mission brief, all prior user commands, Robot Status data, LLM Conversational Responses, situational awareness context data, and a preceding user command (see at least para. [0065] disclose database 518 may store user profile relating to the user 540, and/or the like).
As per claim 8, Gurgu et al. disclose the preceding user command immediately precedes a current user command (see at least [0125] disclose in some embodiment, a new prompt is generated in act 404 using the prior output as a labeled example for the new output. In one embodiment, the new output generated by the machine learning model is different from the prior output).
Claims 11, and 13-15, are system claims corresponding to method claims 3, and 5-7 above. Therefore, they are rejected for the same rationales set forth as above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure:
. Shankar et al. (US 2023/0281422 A1)
. Scherer et al. (US 2022/0092270 A1)
. Ijiri et al. (US 2021/0283771 A1)
. Neuman et al. (US 2019/0102377 A1)
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DALENA TRAN whose telephone number is (571)272-6968. The examiner can normally be reached M-F 7AM-5PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, ADMA MOTT can be reached at 571-270-5376. 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.
/DALENA TRAN/Primary Examiner, Art Unit 3657
Prosecution Insights