DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This Office Action is in response to the submission filed December 23, 2024. Claims 1-20 are pending.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on January 16, 2025; May 29, 2025; June 11, 2026 is being considered by the examiner.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claims 1, 11, and 16 are directed to methods, non-transitory computer readable mediums and systems for generating responses for tech-bio queries. The claims recite limitations for:
identifying, from a client device, a tech-bio query is a data gathering step that can be achieved by a person hearing a tech-bio query;
generating one or more language model prompts from the tech-bio query, wherein the one or more language model prompts comprises one or more descriptions of a plurality of tech-bio exploration tools can be achieved by a person, using pen and paper, creating a list of commands, tasks or purposes related to the query that can be achieved or provide information by a group of people, entities or devices.
generating, utilizing a language machine learning model from one or more language model prompts, an execution request indicating a task for a tech-bio exploration tool of the plurality of tech-bio exploration tools can be achieved by a person, using natural language rules, principles and mathematical algorithms, generate a desired command or task to be executed based on the list of generated prompts.
transmitting the execution request to the tech-bio exploration tool to cause the tech-bio exploration tool to execute the task can be achieved by the person speaking the request or giving a written request to another person.
based on one or more data outputs of the tech-bio exploration tool, utilizing the language machine learning model to generate a response to the tech-bio query can be achieved by the person hearing or reading an output and speaking a response or writing the response on paper for the original requestor.
The recited limitations are directed a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of the generic language model (which can be performed by mathematical calculations), machine learning model (which can be achieved by mathematical calculations), tech-bio tool, computer, system, computer readable medium, and generic computer components. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application because the recited generic language model (which can be performed by mathematical calculations), machine learning model (which can be achieved by mathematical calculations), tech-bio tool, computer, system, computer readable medium, and generic computer components amounts to no more than mere instructions to apply the exception using generic computer components. Accordingly, the elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea. The claims are not patent eligible.
The claim(s) do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, as indicated with respect to integration of the abstract idea into a practical application, the additional elements of the generic language model (which can be performed by mathematical calculations), machine learning model (which can be achieved by mathematical calculations), tech-bio tool, computer, system, computer readable medium, and generic computer components to perform the various steps amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. The claims are not patent eligible.
Dependent claims 2-9, 11-15, and 17-20 do not integrate the judicial exception into a practical application and do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The limitations of the dependent claims are directed to steps of organizing or manipulating functions and commands for generating prompts and requests, steps for repeating the processing for additional prompts or requests; data gathering steps for receiving free-form text, transmitting prompts/requests; and extra-solution activity for displaying outputs. The steps of the dependent claims can be achieved via mental processing, performing mathematical calculations, and/or using pen and paper. The claims are not patent eligible.
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.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Reza et al (US Patent Application Publication No. 2023/0237277), hereinafter Reza, in view of Surdeanu et al (US Patent Application Publication No. 2021/0357585), hereinafter Surdeanu.
Reza discloses developing a set of prompts based on aspects from training data. Regarding claim 1, Reza discloses a computer-implemented method (techniques are provided (e.g., a method, a system, non-transitory computer-readable medium); para [0004]) comprising:
identifying, from a client device, a query [(a query may be used to obtain all text examples within a corpus of data; words may be searched based on the extracted features or a middle word may be replaced with an extracted feature; the service operators may be using one or more client computing devices, which may be portable handheld devices (e.g., an iPhone; Figs. 1-8 & para [0048], [0056], [0088])];
generating one or more language model prompts from the query [(training the machine learning language model on the prompting functions to predict a solution for the task; training a model (e.g., a machine learning language model) to predict a solution for a given task such as sentiment analysis or name entity recognition; a query may be used to obtain all text examples within a corpus of data (e.g., a Wikipedia) pertaining to a given domain; para [0033], [0036], [0048])], wherein the one or more language model prompts comprises one or more descriptions of a plurality of exploration tools [(dynamically developing a contextual set of prompts based on relevant aspects extracted from set of training data; various automated techniques including prompt mining, prompt paraphrasing, a gradient-based search, prompt generation, prompt scoring; Abstract & para [0033], [0073])];
generating, utilizing a language machine learning model from one or more language model prompts [(create prompting functions; training the machine learning language model on the prompting functions to predict a solution for the task; various automated techniques including prompt mining, prompt paraphrasing, a gradient-based search, prompt generation, prompt scoring; para [0048])], an execution request indicating a task for an exploration tool of the plurality of exploration tools [(training the machine learning language model on the prompting functions to predict a solution for the task; various automated techniques including prompt mining, prompt paraphrasing, a gradient-based search, prompt generation, prompt scoring; para [0033], [0048])];
transmitting the execution request to the exploration tool to cause the exploration tool to execute the task [(training the machine learning language model on the prompting functions to predict a solution for the task: para [0033])]; and
based on one or more data outputs of the exploration tool, utilizing the language machine learning model to generate a response to the query [(generated prompting template to create prompting functions; training the machine learning language model on the prompting functions to predict a solution for the task; the set of training data is a corpus of examples for training a model (e.g., a machine learning language model) to predict a solution for a given task such as sentiment analysis or name entity recognition; train a model on the prompting functions to predict a solution for the given task; para [0033], [0036], [0061])].
Reza fails to disclose a tech-bio query and a tech-bio exploration tool. Surdeanu discloses a tech-bio query and a tech-bio exploration tool (rule-based event extraction framework can model underlying syntactic representations of events; system extracts entities (e.g., proteins, other chemicals, biological processes) and events (e.g., biochemical interactions) from literature; the text are shown the output of an entity recognizer; input field accepts the query and produces a search result entities include species, cell lines, organs, cell types, families, cellular or subcellular components, simple chemicals, sites, bioprocesses, and gene or gene products: Figs. 1, 2, 3A, 3B, 4-7, 8A-8D, 9, 10, Abstract & para [0017], [0023], [0077], [0092]).
One having ordinary skill in the art at the time of the invention would have recognized the advantages of implementing the tech-bio exploration tool suggested by Surdeanu, in the system of Reza for the purpose of providing and developing a contextual set of prompts based on relevant aspects extracted from S set of training data and improving prompt-based learning by dynamically developing a contextual set of prompts based on relevant aspects extracted.
Regarding claim 2, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 1. In addition, Reza discloses further comprising utilizing the one or more language model prompts as few-shot learning prompts to enable in-context learning for the language machine learning model from the one or more descriptions of the plurality of tech-bio exploration tools [(conventional large language models pre- trained with prompting demonstrate the ability to infer with the help of few shot learning and can handle a large set of downstream tasks like Q&A, sentiment analysis, NER, etc.; provides a sufficient set of context to learn from a few shot approach (few shot being a prompt with one (1-shot) or more (n-shot, few shot) training examples); para [0025], [0032])].
Regarding claim 3, Reza in view of Surdeanu discloses the computer-implemented method of claim 1. In addition, Reza discloses wherein generating the execution request comprises utilizing the language machine learning model to select the exploration tool from a plurality of exploration tools [(dynamically developing a contextual set of prompts based on relevant aspects extracted from S set of training data; various automated techniques including prompt mining, prompt paraphrasing, a gradient-based search, prompt generation, prompt scoring; Abstract & para [0033], [0073])].
Regarding claim 4, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 3. In addition, Reza discloses further comprising: generating the execution request by generating, utilizing the language machine learning model, a set of instructions to utilize the selected exploration tool to generate the one or more data outputs for the query [(generated prompting template to create prompting functions; training the machine learning language model on the prompting functions to predict a solution for the task; the set of training data is a corpus of examples for training a model (e.g., a machine learning language model) to predict a solution for a given task such as sentiment analysis or name entity recognition; train a model on the prompting functions to predict a solution for the given task; para [0033], [0036], [0061])]; and providing, for display within a graphical user interface of the client device, a representation of the selected exploration tool [(using one or more client computing devices; user interface output devices may include a variety of display devices that visually convey text, graphics and audio/video information; para [0088], [0133])].
Regarding claim 5, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 1. Reza fails to discloses wherein the query comprises free-form text for a request for a bio-tech data output response. Reza fails to disclose a tech-bio query.
Surdeanu teaches the bio=tech query comprises free-form text for a request for a bio-tech data output response [(system extracts entities (e.g., proteins, other chemicals, biological processes) and events (e.g., biochemical interactions) from literature; the text are shown the output of an entity recognizer;(an interactive web-based tool for event grammar development and results visualization: the UI can accept free text to match against, and can be configured to run either a predefined domain grammar or one provided on-the-fly through a text box, allowing for the rapid development and tuning of rules; the web interface is implemented as a client-server web application which runs the EE system on the server and displays the results on the client side; computers typically include known components, such as a processor, input- output controllers, input-output devices, and display devices; para [0017],[0129], [0158])]. It would have been obvious to one of ordinary skill in the art at the time of the invention to include a tech-bio query as taught by the Surdeanu into the system of Reza for the purpose of providing and developing a contextual set of prompts based on relevant aspects extracted from S set of training data and improving prompt-based learning by dynamically developing a contextual set of prompts based on relevant aspects extracted.
Regarding claim 6, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 1. In addition, Reza discloses further comprising, in response to receiving the one or more data outputs from the exploration tool [adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; prompting functions to predict a solution for a task; prompts are in natural language; para [0033], [0038])]:
generating one or more additional language model prompts from the one or more data outputs of the exploration tool [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; prompting functions to predict a solution for a task; prompts are in natural language; para [0033], [0038])]; and
utilizing the language machine learning model to select between generating the response to the query or executing an additional tool from the one or more additional language model prompts (adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038]).
Regarding claim 7, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 6. In addition, Reza discloses further comprising, in response to the language machine learning model selecting to generate the response [(machine learning language model on the prompting functions to predict a solution for a task; a model may be trained on the prompting functions to predict a solution for the given task; pre-training system and further trained on the prompting functions to predict a solution for the given task using pre-training system; Abstract & para [0064])], providing, for display within a graphical user interface of the client device, the response to the query [(using one or more client computing devices; user interface output devices may include a variety of display devices that visually convey text, graphics and audio/video information; para [0088], [0133])], wherein the response comprises at least one of a text output, a visual diagram output, or a data file output [(using one or more client computing devices; user interface output devices may include a variety of display devices that visually convey text, graphics and audio/video information; para [0088], [0133])].
Regarding claim 8, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 6. In addition, Reza discloses further comprising, in response to the language machine learning model selecting to execute the additional tool [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038])]:
generating, utilizing the language machine learning model from the one or more additional language model prompts, an additional execution request indicating an additional task for the additional exploration tool of the plurality of exploration tools [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038])];
transmitting the additional execution request to the additional exploration tool to cause the additional exploration tool to execute the additional task [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; prompting functions to predict a solution for a task; prompts are in natural language; para [0033], [0038])]; and
based on one or more additional data outputs of the additional exploration tool, utilizing the language machine learning model to generate an additional response [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; prompting functions to predict a solution for a task; prompts are in natural language; para [0033], [0038])].
Regarding claim 9, the combination of Reza and Surdeanu teaches the computer-implemented method of claim 1. In addition, Reza discloses further comprising:
generating one or more additional language model prompts from an additional query identified from the client device [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038])];
transmitting an additional execution request generated from the one or more additional language model prompts to an additional tech-bio exploration tool to cause the additional exploration tool to execute an additional task [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038])]; and
based on one or more additional outputs of the additional exploration tool, utilizing the language machine learning model to generate an additional response to the additional query [(adding a prompting pipeline along with pre-training and fine-tuning allows these language models to close the gap and become better learners; utilize the prompt to extract the inherent domain knowledge on a subject area along with the composite knowledge embedded in the language models; prompting functions to predict a solution for a task; prompts are in natural language; para [0025], [0028], [0033], [0038])].
Regarding claims 10-20, claims 10-15 and 16-20 are directed to non-transitory computer readable medium claims and systems, respectively that are rejected under similar rationale as claims 1-9.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANGELA A ARMSTRONG whose telephone number is (571)272-7598. The examiner can normally be reached M,T,TH,F 11:30-8:00.
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, Pierre Desir can be reached at 571-272-7799. 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.
ANGELA A. ARMSTRONG
Primary Examiner
Art Unit 2659
/ANGELA A ARMSTRONG/Primary Examiner, Art Unit 2659
Prosecution Insights