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 .
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they do not include the following reference sign(s) mentioned in the description: Fig. 6 reference character 802 designates Computer Resources, while reference character 602, within the Specification, designates Computer Resources. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim(s) 5 and 15 recites the limitation "the user prompt" in lines 2 and 4. There is insufficient antecedent basis for this limitation in the claim. For the purpose of examination below, “the user prompt” within claims 5 and 15 will be read as “the first prompt”, similar to claims 3 and 13 upon which claims 5 and 15 depend.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Independent claims 1, 11, and 20 recite a method, system, and computer-readable medium (CRM), respectively. These claims therefore invoke a statutory category (machine and process) in Step 1 of the Subject Matter Eligibility Test.
Step 2A, Prong One: Independent claims 1, 11, and 20, under their broadest reasonable interpretation, recite a method, system, and CRM of receiving a first prompt, categorizing the first prompt, altering the first prompt by removing parts of it, finding a historical second prompt similar to the first prompt, ensuring that the historical second prompt provides an outcome that satisfies, and creating a final prompt based on that historical second prompt. These are abstract ideas in the form of certain methods of organizing human activity (i.e. mental processes such as observation, evaluation, judgement, and opinion). The steps of receiving data (i.e. a prompt), categorizing the data, removing parts of the data, finding another data similar, ensuring that the similar data provides a desired outcome, and creating a final data based on the similar data could be performed by a human using pen and paper or by purely mental reasoning.
Step 2A, Prong Two: The claims do not integrate the judicial exception into a practical application. The recitation of “a LLM”, “the LLM”, and “a feedback model” are generic instructions to perform the abstract idea on/using a computer and do not impose a meaningful limit on the judicial exception. The LLM and feedback model are recited at such high-levels of generality and are merely used as tools to perform the abstract idea faster and more efficiently. The data receiving and outputting steps required to perform the method do not add a meaningful limitation. Mere data gathering, analysis, and output do not provide an inventive concept. There is no improvement to the functioning of prompting an LLM, prompt optimization techniques, the functioning of the computer, LLM, or feedback model themselves, or to any other technology or technical field.
Step 2B: The claims do not include any additional elements that amount to significantly more than the judicial exception. The only additional element beyond the abstract idea is the LLM, which performs generic computational functions such as receiving, analyzing, and outputting data. Such elements are well-understood, routine, and conventional within the field.
Accordingly, claims 1, 11, and 20 are directed to an abstract idea and do not include significantly more than the abstract idea itself.
With respect to claims 2 and 12, the claims relate to rephrasing the first prompt based on the historical prompt. This is a mental process that could be performed by a human using pen and paper or by purely mental reasoning. No additional elements are present.
With respect to claims 3 and 13, the claims relate to rephrasing the first prompt using placeholders, generating a second prompt with the placeholders, and asking a user to obtain further information. This is a mental process that could be performed by a human using pen and paper or by purely mental reasoning. No additional elements are present.
With respect to claims 4 and 14, the claims relate to rephrasing the first prompt using few-shot learning or prompt tuning to obtain the placeholders. These are processes utilized by models and are thereby treated as generic instructions to perform the abstract idea on/using a computer and do not impose a meaningful limit on the judicial exception. No additional elements are present.
With respect to claims 5 and 15, the claims relate to rephrasing the first prompt and rephrasing the second prompt based on a score based on a feedback model. This is a mental process that could be performed by a human using pen and paper or by purely mental reasoning. The only additional element is “the feedback model”, which is a generic instruction to perform the abstract idea on/using a computer and does not impose a meaningful limit on the judicial exception. No additional elements are present.
With respect to claims 6 and 16, the claims relate to the prompt categorization, prompt preprocessing, and retrieving and identifying of the historical prompts. These actions are mental processes that could be performed by a human using pen and paper or by purely mental reasoning. The only additional elements are “a first software engine”, “a second software engine”, and “a third software engine”, which are generic instructions to perform the abstract idea on/using a computer and do not impose a meaningful limit on the judicial exception. No additional elements are present.
With respect to claims 7-8 and 17, the claims relate to the prompt being input by either a human or robotic user. This is insignificant extra-solution activity; pre-solutional activities do not provide an inventive concept. No additional elements are present.
With respect to claims 9 and 18, the claims relate to receiving an output from the LLM after inputting the final prompt. This is insignificant extra-solution activity; generic data outputting does not provide an inventive concept. The only additional element is “the LLM”, which is a generic instruction to perform the abstract idea on/using a computer and does not impose a meaningful limit on the judicial exception. No additional elements are present.
With respect to claims 10 and 19, the claims relate to providing the LLM output responsive to the final prompt back to the user. This is insignificant extra-solution activity; generic data outputting and generic data transmission do not provide an inventive concept. No additional elements are present.
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) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Figueredo de Santana et al. (US Patent No. 12,619,822), hereinafter referred to as Santana, in view of Cuomo et al. (US Patent No. 12,455,980), hereinafter referred to as Cuomo.
Regarding claim 1, Santana discloses a method of optimizing utilization of large language models (LLMs), the method comprising, by a computer system: receiving a first prompt for a LLM (Santana col. 5 lines 30-35);
retrieving one or more historical prompts (Santana col. 13 lines 41-56) for the LLM based on a determination that the one or more historical prompts are similar to the first prompt (Santana col. 14 lines 66-67 through col. 15 lines 1-12);
identifying, based on a feedback model, at least one historical prompt of the one or more historical prompts as producing an accurate response from the LLM (Santana col. 13 lines 41-56);
generating a final prompt for the LLM based on the at least one historical prompt (Santana Fig. 8 reference character 818, specifically "using the adjusted first prompt" implies that a final prompt has been generated);
and providing the final prompt to the LLM (Santana Fig. 8 reference character 818, specifically "cause the model to produce a first content" implies that the final prompt is being input to the model).
However, Santana fails to disclose categorizing the first prompt into a category based on a topic modeling algorithm; processing the first prompt to remove information determined to be irrelevant to the category. Cuomo teaches a privacy preservation system and method for large language models.
Cuomo teaches categorizing the first prompt into a category based on a topic modeling algorithm (Cuomo Fig. 4 reference character 406);
processing the first prompt to remove information determined to be irrelevant to the category ("Next at block 404, the method 400 includes generating pre-processed prompt data. In some embodiments, the pre-processing module 304 prepares the prompt data 206 received from the user device 204 for topic modeling. The pre-processing module 304 removes personally identifiable information, such as the name, address, identification number (e.g., social security number, driver's license number, etc.). In some embodiments, removing personally identifiable information involves using regular expressions or another method to detect and remove personally identifiable information," Cuomo col. 13 lines 48-58 and "With respect to any flowcharts, depending upon the technology involved, the operations can be performed in a different order than what is shown in a given flowchart. For example, again depending upon the technology involved, two operations shown in successive flowchart blocks may be performed in reverse order, as a single integrated step, concurrently, or in a manner at least partially overlapping in time," Cuomo col. 4 lines 7-14).
It would have been obvious to one of ordinary skill within the art before the effective filing date of the claimed invention to have modified Santana’s method of using historical prompts to alter and generate a final, refined prompt by including Cuomo’s teaching of utilizing categorization of prompts and removing unnecessary information. Categorizing the prompts would decrease the computational effort and time necessary to retrieve similar historical prompts, as similar prompts would already be consolidated into categories. This is a well-known technique within the art. Removing unnecessary information irrelevant to the category, through tokenization or the like, is also known well-known within the art and would have been an obvious inclusion, as it also decreases computational effort and time.
Regarding claim 2, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses rephrasing the first prompt based on the at least one historical prompt, wherein the final prompt for the LLM is generated based on the rephrased first prompt (Santana Fig. 8 reference character 804).
Regarding claim 3, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses rephrasing the first prompt based on the at least one historical prompt, wherein the rephrasing comprises inserting one or more placeholders for additional information (Santana Fig. 7 reference character 410 and 720);
generating a second prompt based on the at least one historical prompt, the second prompt comprising the one or more placeholders for additional information (Santana Fig. 7 reference character 410);
and interacting with a user to receive the additional information, wherein the generation of the final prompt comprises regenerating the second prompt to include the received additional information in place of the one or more placeholders (Santana Fig. 7 reference character 730 requesting or recommending more information).
Regarding claim 4, Santana, in view of Cuomo, discloses all of the limitations of claim 3. Santana further discloses wherein the rephrasing further comprises applying at least one of few-shot learning or prompt tuning to yield at least one of the one or more placeholders for additional information ("Optionally, data of a particular prompt in the plurality of prompts includes content output by a model in response to the prompt, and additional data or metadata, such as a specification of the model to be used, one or more parameters that adjust the model's output (e.g., temperature, a measure of variability of the model's output in response to the same prompt), or one or more operators adjusting the weight to be given to particular subgoals within a prompt," Santana col. 13 lines 48-56).
Regarding claim 5, Santana, in view of Cuomo, discloses all of the limitations of claim 3. Santana further discloses wherein the rephrasing comprises: rephrasing the user prompt based on a first historical prompt of the plurality of historical prompts to yield a first rephrased user prompt (Santana Fig. 7 reference character 410 and 720);
and rephrasing the user prompt based on a second historical prompt of the plurality of historical prompts to yield a second rephrased user prompt (Santana Fig. 4 shows two prompts being used to create a template, i.e. rephrasing the user prompt), wherein the generating the second prompt comprises selecting the first rephrased user prompt or the second rephrased user prompt based on a scoring relative to the feedback model (Santana Fig. 8 reference character 808).
Regarding claim 6, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses and the retrieving and the identifying are performed via a third software engine configured for evaluating prompts relative to the plurality of historical prompts (Santana Fig. 3 reference character 310).
However, Santana fails to disclose wherein: the categorizing is performed via a first software engine configured for category selection; the processing is performed via a second software engine configured for prompt preprocessing.
Cuomo teaches wherein: the categorizing is performed via a first software engine configured for category selection (Cuomo Fig. 3 reference character 306);
the processing is performed via a second software engine configured for prompt preprocessing (Cuomo Fig. 3 reference character 304).
It would have been obvious to one of ordinary skill within the art before the effective filing date of the claimed invention to have modified Santana’s method of using historical prompts to alter and generate a final, refined prompt by including Cuomo’s teaching of utilizing categorization of prompts and removing unnecessary information. Categorizing the prompts would decrease the computational effort and time necessary to retrieve similar historical prompts, as similar prompts would already be consolidated into categories. This is a well-known technique within the art. Removing unnecessary information irrelevant to the category, through tokenization or the like, is also known well-known within the art and would have been an obvious inclusion, as it also decreases computational effort and time.
Regarding claim 7, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses wherein the first prompt is input by a human user (Santana col. 5 lines 30-35, while "user" is not specified to be human or robotic, it is commonly known within the art for humans to interact with large language models or other models).
Regarding claim 8, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses wherein the first prompt is input by a robotic user (Santana col. 5 lines 30-35, while "user" is not specified to be human or robotic, it is commonly known within the art to use virtual agents or robots to enter prompts into models).
Regarding claim 9, Santana, in view of Cuomo, discloses all of the limitations of claim 1. Santana further discloses receiving an output from the LLM responsive to the providing the final prompt ("Application 300, using the new prompt or an adjusted version of the new prompt, causes a model to product a content. One implementation of application 300 stores a user's response to a recommendation, for use in improving future recommendations," Santana col. 15 lines 63-67).
Regarding claim 10, Santana, in view of Cuomo, discloses all of the limitations of claim 9. Santana further discloses providing information related to the output to a user associated with the first prompt ("Application 300, using the new prompt or an adjusted version of the new prompt, causes a model to product a content. One implementation of application 300 stores a user's response to a recommendation, for use in improving future recommendations," Santana col. 15 lines 63-67).
As to claims 11-19, system claims 11-19 and method claims 1-10 are related as method and system of using same, with each claimed element’s function corresponding to the respective method step. Accordingly, claims 11-19 are similarly rejected under the same rationale as applied above with respect to the method claims (claim 17 is a combination of claims 7 and 8, and is therefore rejected under the same rationale as well).
As to claim 20, computer-readable medium (CRM) claim 20 and method claim 1 are related as method and CRM of using same, with each claimed element’s function corresponding to the method step. Accordingly, claim 20 is similarly rejected under the same rationale as applied above with respect to the method claim.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
US Patent Application Publication No. 2026/0057256
Agarwal et al., “PROMPTWIZARD: TASK-AWARE PROMPT OPTIMIZATION FRAMEWORK”, 10/03/2024
Juneja et al., “Task Facet Learning: A Structured Approach to Prompt Optimization”, 06/15/2024
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ADAM MICHAEL WEAVER whose telephone number is (571)272-7062. The examiner can normally be reached Monday-Friday, 8AM-5PM EST.
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.
/ADAM MICHAEL WEAVER/Examiner, Art Unit 2658
/RICHEMOND DORVIL/Supervisory Patent Examiner, Art Unit 2658
Prosecution Insights