DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 21 November 2025 has been entered. Claims 1-20 have been examined and are pending.
Priority
Acknowledgment is made of applicant's claim for foreign priority based on an application filed in China on 15 April 2024.
Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
US 20240394285
Abstract
Answering a user query by generating multiple prompts as LLM module inputs
US 20240346256
Par. 46
Generating subsequent augmented prompts based on a user query, product information, previous generated prompts, etc.
US 20240311407
Pars. 18-26, 30, 72
Providing product recommendations by converting user queries to LLM prompts
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 1-20 are directed to one of the eligible categories of subject matter.
With respect to independent claims 1, 8 and 15, the converting, generating, processing, search, call, and executing cover performance of the limitations manually and/or in the mind (mental processes abstract idea). The receiving, and providing are recited at a high level of generality and do not add meaningful limitations to the abstract idea; these limitations are directed to insignificant extra solution activities. The claims as a whole merely describe how to generally “apply” the exception in a computer environment using generic computer functions or components such as the claimed processor, memory, and medium. Even when viewed in combination, these additional 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 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claims are not patent eligible.
With respect to dependent claims 2-7, 9-14 and 16-20, the include and provide limitations are recited at a high level of generality and do not add meaningful limitations to the abstract idea. The claims as a whole merely describe how to generally “apply” the exception in a computer environment using generic computer functions or components. Even when viewed in combination, these additional 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 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al., Pub. No.: US 20240241897 A1, hereinafter Wang, in view of NPL Schick et al., Toolformer: Language Models Can Teach Themselves to Use Tools, hereinafter Schick, and further in view of J. Ordorica de la Torre, Pub. No.: US 20250139683 A1, hereinafter Ordorica de la Torre.
As per claim 1, Wang discloses A method comprising:
receiving, by a processor, a query for a product recommendation (pars. 97-98);
converting the query into a first prompt for a large language model, wherein the first prompt confines the product recommendation to a domain (par. 99 discloses a first prompt as claimed: “Show relevant products for ‘milk,’” wherein the recommendation is confined to the domain of milk related products and/or food; par. 55, 80, 82 disclose categories and types as examples of domains; also, pars. 41, 86, 111 disclose that the search and recommendation is domain-specific);
generating a second prompt to guide the large language model in marking content with a task delimiter […] for searching auxiliary data according to the query (pars. 99, 116, 119 disclose multiple additional prompts such as “Show replacements for [P001]. List with item identifiers,” each prompt marking content with task delimiters such as item identifier “[P001]” which is a marking of content P001 with task delimiter “[ ]”, task identifiers, annotations, tokenization, etc., wherein item identifiers, task identifiers, annotations, tokens, etc., as per pars. 105-116, are used for searching at least an online catalog (i.e. for searching auxiliary data according to the query));
providing the first prompt and the second prompt to the large language model to generate a task sequence, wherein the task sequence includes a task for the searching of the auxiliary data in real time based on the task delimiter (see mapping above including pars. 116-122; at least pars. 5, 24, 71-76 indicate that all the steps occur in real time), and
Wang, par. 116 discloses tasks including search retrieval, item replacement, and personalized search suggestion in connection with an online catalog and pars. 78-82, 105, 108, 124 further describe searching the online catalog; Wang does not expressly disclose, however, Yao in the related field of endeavor of information retrieval discloses and a task keyword (Schick, fig. 1, different API’s are called from within LLM prompts and are written as task keywords inside delimiter brackets, the API call keywords corresponding to network services’ external tools such as calculator, a Q&A system, a search engine, a translation system, and a calendar), … wherein the task for the searching of the auxiliary data includes a network service call to perform a search based on the task keyword (Schick, see rejection of previous limitation for at least a search engine); processing the task sequence to determine the task that includes the network service call (Schick, see rejection of previous limitations including section 2, Executing API Calls and section 4.4 first par.).
executing the task to obtain the auxiliary data (see mapping above including Wang pars. 116-122 and Schick as cited), wherein the task includes initiating the network service call (see Schick as cited above);
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the cited references because Schick would have allowed Wang to “use external tools via simple APIs… incorporate a range of tools, including a calculator, a Q&A system, a search engine, a translation system, and a calendar … achieves substantially improved zero-shot performance across a variety of downstream tasks, often competitive with much larger models, without sacrificing its core language modeling abilities.” (Schick, abstract.)
providing a third prompt to the large language model to generate […] data in response to the third prompt based on the auxiliary data (Wang par. 123 makes it clear that the personalization uses an LLM and prompts the LLM, Wang par. 124 makes it clear that the personalized search LLM’s predicted items undergo one or more additional rounds of personalization (i.e. third prompt, fourth prompt as claimed) before final recommended items are selected);
The combination does not expressly disclose, however Ordorica de la Torre, in the related field of endeavor of information retrieval discloses a summary of the auxiliary data in pars. 37, 115, 117-120, 125, 131, 139.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the cited references because Ordorica de la Torre’s teaching would have allowed the combination to implement the well-known technique of prompting an LLM to generate narrative descriptions and summaries of data. Both references are directed to providing end users with product recommendations using machine learned models.
providing a fourth prompt with candidate products based on the summary to the large language model to generate the product recommendation in response to the fourth prompt (see Wang as cited in the rejection of the previous limitation for a disclosure of fourth prompt and par. 124 “the item predictions 845 may undergo one or more additional rounds of personalization” (i.e. third prompt, fourth prompt as claimed) and see Ordorica de la Torre as cited above and rationale to combine); and
providing the product recommendation in response to the query (see Wang at least fig. 10 and Ordorica de la Torre at least par. 37).
As per claim 2, Wang in view of Schick and Ordorica de la Torre discloses the method of claim 1, wherein the fourth prompt further includes a criterion (see rejection of last 3 limitations of claim 1 – note that every LLM based personalization round in Wang includes multiple instances of different “criterion” such as a prompt, query and/or recommended item content, task identifiers, annotations, tokenization, etc.).
As per claim 3, Wang in view of Schick and Ordorica de la Torre discloses The method of claim 1, wherein the third prompt provided to the large language model includes text content from the auxiliary data (see rejection of the third prompt limitation in claim 1).
As per claim 4, Wang in view of Schick and Ordorica de la Torre discloses the method of claim 1, further comprising providing the product recommendation as an answer to the query (see Wang at least fig. 10 and Ordorica de la Torre at least par. 37).
As per claim 5, Wang in view of Schick and Ordorica de la Torre discloses the method of claim 1, wherein the fourth prompt includes a criterion from the summary (see rejection of the fourth prompt limitation in claim 1 including Ordorica de la Torre pars. 37, 115, 117-120, 125, 131, 139 and above provided rationale to combine).
As per claim 6, Wang in view of Schick and Ordorica de la Torre discloses the method of claim 1, wherein the first prompt includes a persona (Wang, at least pars. 60, 99, 116, 124).
As per claim 7, Wang in view of Schick and Ordorica de la Torre discloses The method of claim 1, wherein the first prompt includes an inference context based on an implicit requirement of the query (Wang par. 99 wherein the prompt infers that a relevant product is desired based on what is implied by the query of par. 98).
As per claim 8, Wang discloses An information handling system, comprising: a processor; and a memory storing instructions that when executed cause the processor to perform operations including:
receiving a query for a product recommendation (pars. 97-98);
converting the query into a first prompt for a large language model, wherein the first prompt confines the product recommendation to a domain (par. 99 discloses a first prompt as claimed: “Show relevant products for ‘milk,’” wherein the recommendation is confined to the domain of milk related products and/or food; par. 55, 80, 82 disclose categories and types as examples of domains; also, pars. 41, 86, 111 disclose that the search and recommendation is domain-specific);;
generating a second prompt to guide the large language model in marking content with a task delimiter […] for searching auxiliary data according to the query (pars. 99, 116, 119 disclose multiple additional prompts such as “Show replacements for [P001]. List with item identifiers,” each prompt marking content with task delimiters such as item identifier “[P001]” which is a marking of content P001 with task delimiter “[ ]”, task identifiers, annotations, tokenization, etc., wherein item identifiers, task identifiers, annotations, tokens, etc., as per pars. 105-116, are used for searching at least an online catalog (i.e. for searching auxiliary data according to the query));
providing the first prompt and the second prompt to the large language model to generate a task sequence, wherein the task sequence includes a task for the searching of the auxiliary data in real time based on the task delimiter (see mapping above including pars. 116-122; at least pars. 5, 24, 71-76 indicate that all the steps occur in real time), and
Wang, par. 116 discloses tasks including search retrieval, item replacement, and personalized search suggestion in connection with an online catalog and pars. 78-82, 105, 108, 124 further describe searching the online catalog; Wang does not expressly disclose, however, Yao in the related field of endeavor of information retrieval discloses and a task keyword (Schick, fig. 1, different API’s are called from within LLM prompts and are written as task keywords inside delimiter brackets, the API call keywords corresponding to network services’ external tools such as calculator, a Q&A system, a search engine, a translation system, and a calendar), … wherein the task for the searching of the auxiliary data includes a network service call to perform a search based on the task keyword (Schick, see rejection of previous limitation for at least a search engine); processing the task sequence to determine the task that includes the network service call (Schick, see rejection of previous limitations including section 2, Executing API Calls and section 4.4 first par.).
executing the task to obtain the auxiliary data (see mapping above including Wang pars. 116-122 and Schick as cited), wherein the task includes initiating the network service call (see Schick as cited above);
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the cited references because Schick would have allowed Wang to “use external tools via simple APIs… incorporate a range of tools, including a calculator, a Q&A system, a search engine, a translation system, and a calendar … achieves substantially improved zero-shot performance across a variety of downstream tasks, often competitive with much larger models, without sacrificing its core language modeling abilities.” (Schick, abstract.)
providing a third prompt to the large language model to generate […] data in response to the third prompt based on the auxiliary data (Wang par. 123 makes it clear that the personalization uses an LLM and prompts the LLM, Wang par. 124 makes it clear that the personalized search LLM’s predicted items undergo one or more additional rounds of personalization (i.e. third prompt, fourth prompt as claimed) before final recommended items are selected).
The combination does not expressly disclose, however Ordorica de la Torre, in the related field of endeavor of information retrieval discloses a summary of the auxiliary data in pars. 37, 115, 117-120, 125, 131, 139.
Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of the cited references because Ordorica de la Torre’s teaching would have allowed the combination to implement the well-known technique of prompting an LLM to generate narrative descriptions and summaries of data. Both references are directed to providing end users with product recommendations using machine learned models.
receiving the product recommendation from the large language model based on candidate products included in the summary in response to providing a fourth prompt with the candidate products to the large language model (see Wang as cited in the rejection of the previous limitation for a disclosure of fourth prompt and fig. 10 and par. 124 “the item predictions 845 may undergo one or more additional rounds of personalization” (i.e. third prompt, fourth prompt as claimed) and see Ordorica de la Torre as cited above and rationale to combine); and
providing the product recommendation in response to the query (see Wang at least fig. 10 and Ordorica de la Torre at least par. 37).
As per claims 9-20, they are analogous to claims above and therefore likewise rejected.
Response to Arguments
Applicant's arguments filed 21 November 2025 have been fully considered.
With respect to the 35 USC 101 rejection, the remarks present that the rejection is moot in light of the amendments. Examiner respectfully disagrees. Confining a prompt to a domain and using a task keyword as claimed are mental processes and can be done with pen and paper (by writing a query having a certain scope, destination and/or task on paper). The in-real time searching limitation merely describes a desired result of a search, it does not change how a computer or network operates.
With respect to the prior art rejection, NPL Schick et al., Toolformer: Language Models Can Teach Themselves to Use Tools, has been applied in response to claim amendments.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SYED HASAN whose telephone number is (571)270-5008. The examiner can normally be reached M-F 8am - 5 pm.
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, Boris Gorney can be reached at (571)270-5626. 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.
/SYED H HASAN/Primary Examiner, Art Unit 2154
Prosecution Insights