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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 02/20/2026 and 04/29/2026 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Drawings
The drawings were submitted on 12/06/2024. These drawings are reviewed and accepted by the examiner.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-18 and 20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Ramnath et al. (US 20260119894 A1).
Regarding claims 1 and 20, Ramnath teaches:
“causing a generative model query of a user to be analyzed to determine whether retrieval augmented generation (RAG) should be used to generate a response to the generative model query” (par. 0070; ‘In some examples, the system may determine whether to perform the RAG 422 or to perform the direct generation 424. In the RAG 422, for each query, the system may retrieve possible candidates upon which the response may be based.’);
“in response to a determination that RAG should be used, causing to be assembled, into a generative model input prompt, data indicative of the generative model query, as well as data indicative of one or both of: (i) user-specific conditioning data associated with the user, and (ii) personal RAG data of the user comprising one or more past user interactions between the user and one or more computing devices, wherein the user-specific conditioning data was built over time based at least in part on the personal RAG data of the user” (par. 0070; ‘In either case, the adapter 4330 that was finetuned on the tenant's specific conversations, knowledge base, articles, templates, or other tenant-associated information, the direct generation may be employed, and may, in some cases, involve reduced latency due to avoiding any bottlenecks that may be associated with the RAG 422.’); and
“causing the generative model input prompt to be processed using one or more generative models to generate generative model output that comprises the response to the generative model query, and that is conditioned on one or both of the user-specific conditioning data or the personal RAG data of the user” (par. 0071; ‘Such merging may generate or otherwise result in the merged model 426, which may perform the response generation 432, which may involve generative AI model processing of the query to generate the response.’).
Regarding claim 2 (dep. on claim 1), Ramnath further teaches:
“wherein the generative model query is processed using one or more machine learning models trained to generate output indicative of whether RAG should be used” (par. 0026; ‘Additionally, or alternatively, the system 100 may support the use of a large language model (generative AI model), such as the generative AI component 145. In some examples, a generative AI component 145 may also be referred to as any of an artificial intelligence (AI), a generative AI (GAI), a GAI model, a large language model (LLM).’).
Regarding claim 3 (dep. on claim 2), Ramnath further teaches:
“wherein the output comprises an indication that either the user-specific conditioning data or the personal RAG data of the user should be assembled into the generative model input prompt” (par. 0056; ‘For example, the user feedback 322 may indicate one or more portions of the transcripts 328 that may include a preferred generation 326 or a dispreferred generation 334.’).
Regarding claim 4 (dep. on claim 2), Ramnath further teaches:
“wherein the one or more machine learning models comprises one or more of the generative models” (par. 0046; ‘In some examples, the server 210 may pass the query 265 to the generative AI model 215, which may generate the response 270 (e.g., based on the merged parameters 260) and the server 210 may provide the response 270 to the client 205.’).
Regarding claim 5 (dep. on claim 4), Ramnath further teaches:
“wherein a first generative model is used to process the generative model query and a second generative model different from the first generative model is used to process the generative model input prompt” (par. 0082; ‘At 530, the server 515 may merge the first set of parameters of the second generative AI model with a second set of parameters associated with a base model of the second generative AI model to generate a merged set of parameters.’).
Regarding claim 6 (dep. on claim 5), Ramnath further teaches:
“wherein the first generative model has fewer parameters than the second generative model” (par. 0082; ‘In some examples, to merge the first set of parameters with the second set of parameters, the server 515 may apply a weight update to the second set of parameters of the second generative AI model and the weight update is based on the first set of parameters.’).
Regarding claim 7 (dep. on claim 5), Ramnath further teaches:
“wherein the first generative model is a student model and the second generative model is a teacher model” (par. 0071; ‘In either case (e.g., involving the RAG 422 or the direct generation 424) the adapter 430 for the tenant may be merged with the base model 428 (e.g., parameters associated with the adapter 430 may be merged with parameters of the base model 428, such as via LoRA or PEFT techniques). Such merging may generate or otherwise result in the merged model 426, which may perform the response generation 432, which may involve generative AI model processing of the query to generate the response.’).
Regarding claim 8 (dep. on claim 4), Ramnath further teaches:
“wherein the same generative model is used to process the generative model query and to process the generative model input prompt” (par. 0082; ‘In some examples, the first generative AI model and the second generative AI model are a same generative AI model.’).
Regarding claim 9 (dep. on claim 4), Ramnath further teaches:
“causing to be assembled, as a RAG analysis input prompt, data indicative of the generative model query, wherein one or more of the generative models is used to process the RAG analysis input prompt to generate the output indicative of whether RAG should be used” (par. 0070; ‘In some examples, the system may determine whether to perform the RAG 422 or to perform the direct generation 424. In the RAG 422, for each query, the system may retrieve possible candidates upon which the response may be based’).
Regarding claim 10 (dep. on claim 9), Ramnath further teaches:
“wherein the RAG analysis input prompt is further assembled to include the user-specific conditioning data and/or one or more of the past user interactions forming the personal RAG data of the user” (par. 0070; ‘In either case, the adapter 4330 that was finetuned on the tenant's specific conversations, knowledge base, articles, templates, or other tenant-associated information, the direct generation may be employed, and may, in some cases, involve reduced latency due to avoiding any bottlenecks that may be associated with the RAG 422.’).
Regarding claim 11 (dep. on claim 1), Ramnath further teaches:
“in response to a determination that RAG should not be used, refraining from assembling the data indicative of the user-specific conditioning data or personal RAG data of the user into the generative model prompt” (par. 0070; ‘Additionally, or alternatively, in the direct generation 424, the system may directly generate the output without performing the RAG 422.’).
Regarding claim 12 (dep. on claim 1), Ramnath further teaches:
“wherein the user-specific conditioning data comprises a summary of the user generated using the personal RAG data of the user, and wherein the summary comprises a textual summary or one or more embeddings” (par. 0040; ‘For example, such technique may allow for transfer learning, where a generative AI model finetuned according to the techniques described herein may also be used to augment other generative AI model use cases such as case summarization, knowledge creation, or other techniques for a given tenant as a result of the generative AI model being trained with the tenant-specific knowledge and jargon.’).
Regarding claim 13 (dep. on claim 1), Ramnath further teaches:
“wherein one or more of the user interactions comprises one or more of: electronic correspondence sent or received by the user using one or more of the computing devices; a document accessed by the user using one or more of the computing devices; a software application installed on one or more of the computing devices and used by the user; a change to an installed software application on one or more of the computing devices and used by the user; a change made to a software application settings or functionality on one or more of the computing devices and used by the user; a change made to a computing device configuration on one or more of the computing devices and used by the user; a change made to a security or privacy configuration of a resource controlled by the user; one or more digital images captured or altered by the user; one or more content purchases by the user; one or more preferences provided explicitly by the user; rejection of generative model output provided to the user based on the user-specific conditioning data; one or more social media posts of the user; one or more location trajectories accumulated by one or more of the computing devices; or one or more readings from one or more physiological sensors worn by the user” (par. 0054; ‘For example, the reference articles 320, the transcripts 328, or both, may be analyzed to determine user feedback on the transcripts 328 of actual chat interactions.’).
Regarding claim 14 (dep. on claim 1), Ramnath further teaches:
“wherein the one or more new user interactions comprise one or more of: commissioning a new smart appliance into a coordinated ecosystem of smart appliances associated with the user; altering a configuration of a smart appliance within the coordinated ecosystem; or decommissioning a smart appliance from the coordinated ecosystem” (par. 0020; ‘This may include support for sales, service, marketing, community, analytics, applications, and the Internet of Things.’ The support for the Internet of Things reads on commissioning smart appliances, altering configurations, etc.).
Regarding claim 15 (dep. on claim 1), Ramnath further teaches:
“wherein the analysis of the generative model query is performed at a resource-constrained edge device” (par. 0036; ‘Additionally, or alternatively, the generative AI model may be limited by the quantity of tokens it can process in each turn. For example, it may not be possible to feed in an entire knowledge base of an organization in the context of each reply in a conversation.’).
Regarding claim 16 (dep. on claim 1), Ramnath further teaches:
“wherein the personal RAG data is retrieved based on the user-specific conditioning data” (par. 0072; ‘In at least these ways, a generative AI model may utilize the tenant-specific or tenant-associated information to respond to the query, thereby improving RAG techniques and overcoming obstacles with generative AI model usage, particularly when domain-specific knowledge is desirable while still maintaining security and isolation considerations between different tenants.’).
Regarding claim 17 (dep. on claim 16), Ramnath further teaches:
“wherein the personal RAG data is retrieved based on: one or more mappings between the user-specific conditioning data and one or more data sources that store at least a portion of the personal RAG data; a query generated using one or more of the generative models, wherein the query is generated by conditioning one or more of the generative models using the user-specific conditioning data; or a semantic similarity search” (par. 0072; ‘In at least these ways, a generative AI model may utilize the tenant-specific or tenant-associated information to respond to the query, thereby improving RAG techniques and overcoming obstacles with generative AI model usage, particularly when domain-specific knowledge is desirable while still maintaining security and isolation considerations between different tenants.’).
Regarding claim 18 (dep. on claim 1), Ramnath further teaches:
“wherein the personal RAG data is limited to user interactions during a predetermined time interval” (par. 0056; ‘For example, the user feedback 322 may indicate one or more portions of the transcripts 328 that may include a preferred generation 326 or a dispreferred generation 334.’).
Allowable Subject Matter
Claim 19 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter:
The prior art of record, whether taken alone or in combination, fails to teach, inter alia, analyzing the response to the query, as in “causing data indicative of the response to the generative model query to be analyzed to determine whether RAG should be used to generate an augmented response to the generative model query; in response to a determination that RAG should be used to generate the augmented response to the generative model response, causing to be assembled, into a new generative model input prompt, data indicative of the generative model response, as well as data indicative of one or both of: (i) the user-specific conditioning data associated with the user, and (ii) personal RAG data of the user; and causing the new generative model input prompt to be processed using one or more of the generative models to generate updated generative model output that comprises the augmented response to the generative model query.”
Conclusion
Other pertinent prior art are cited in the PTO-892 for the applicant's consideration.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARK VILLENA whose telephone number is (571)270-3191. The examiner can normally be reached 10 am - 6pm EST Monday through Friday.
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.
MARK . VILLENA
Examiner
Art Unit 2658
/MARK VILLENA/Examiner, Art Unit 2658
Prosecution Insights