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 .
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) 1-5 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Taveira et al. (U.S. Patent Application Pub. No. 2025/0348910, hereinafter “Taveira”), in view of Deutsch et al. (U.S. Patent Application Pub. No. 2025/0384222, hereinafter “Deutsch”).
In regard to claim 1, Taveira discloses a method of automated pre-prompt generation (Fig. 7, 700), the method comprising:
receiving, by a user device, an indication of at least one user preference for a user (user preferences 714, paragraph [0103]), the at least one user preference indicative of at least one first characteristic, preferred by a user, of natural-language outputs generated by a machine-learning language model based on user-provided natural-language text inputs (user preferences include, for example, advertisement preferences for natural language responses, paragraph [0089]);
receiving, by a server and from the user device, a natural-language text prompt provided by the user to a chat application operating on the user device (a prompt 710 is received from a user 702, paragraph [0103]);
receiving, by the server and from the user device, the at least one user preference (a user profile comprising user preferences is provided from a user’s device, paragraphs [0091] and [0103]);
receiving at least one first operator preference indicative of at least one second characteristic, preferred by an operator of the server, of the natural-language outputs (brand/advertiser preferences 716 are received, paragraph [0103]);
modifying, by the server, a prompt for the machine-learning language model based on the received at least one user preference and the at least one first operator preference to generate a modified system prompt (ad module 750 modifies prompt 710 based on user preferences 714 and brand/advertiser preferences 716, paragraph [0103]);
providing, by the server, the modified prompt as input to the machine-learning language model to generate a natural-language text output (updated prompts 720 are provided to a generative language model system 704, paragraph [0107]);
transmitting, by the server, the natural-language text output to the user device (output 722 is provided to the user device, see Fig. 7 and paragraph [0107]); and
communicating, by the chat application via the user device, the natural-language text output to the user (see Fig. 5, a response 506 is provided to a user 502 in a chat interface, paragraphs [0092-0094]).
While Taveira discloses modifying a natural language prompt according to user preferences and operator preferences, Taveira does not expressly disclose first providing a modified system prompt, then subsequently providing the natural-language text prompt provided by the user as a separate prompt.
Deutsch discloses a method for providing custom model instructions based on user preferences, comprising providing, by a server, a modified system prompt as an initial input to a machine-learning language model (in response to receiving a natural language prompt from a user, paragraph [0061]; custom instructions related to a user’s preferences are provided to a natural language model as a system prompt, paragraphs [0067] and [0071-0072]); and
providing, by the server and after providing the modified system prompt, the natural-language text prompt as an input to the machine-learning language model to generate a natural-language text output (the natural language prompt from the user is also provided to the natural language model to generate a natural language response, paragraphs [0067] and [0071-0072]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to split the prompt of Taveira into an initial system prompt, followed by the natural language text prompt, because providing custom model instructions as a system prompt prefix before a user input improves the accuracy, efficiency, trainability, and generation of language model responses based on custom instructions, as taught by Deutsch (paragraphs [0005-0006]).
In regard to claim 2, Taveira discloses receiving the prompt comprises receiving a user identifier corresponding to the user (user profile, paragraph [0092]).
In regard to claim 3, Taveira discloses receiving the at least one first operator preference comprises querying, by the server, a first database using the user identifier to retrieve the at least one first operator preference (the brand preferences are selected using the user profile, paragraph [0103]).
In regard to claim 4, Taveira discloses receiving at least one second operator preference indicative of at least one third characteristic, preferred by the operator of the server, of the natural-language outputs, wherein modifying the system prompt comprises modifying the system prompt based on the on the received at least one user preference, the at least one first operator preference, and the at least one second operator preference to generate the modified system prompt (brand preferences 716 additionally include ad keywords, ad emotions and advertisement data, all of which are used in modifying the prompt, paragraphs [0103-0107]).
In regard to claim 5, Taveira discloses receiving the at least one second operator preference comprises querying, by the server, a second database using the user identifier to retrieve the at least one second operator preference (the brand/ad term is provided from different sources, paragraph [0103]).
In regard to claim 20, Taveira discloses a system for language generation, the system comprising:
a user device (a user prompt provided from a device to a server, paragraph [0085]) comprising:
a first processor (general purpose processor, paragraph [0228]); and
at least one first memory (RAM, paragraph [0229]) storing at least one user preference (user preferences 714, paragraph [0103]) indicative of at least one first characteristic, preferred by a user, of natural-language outputs generated by a machine-learning language model based on user-provided natural-language text inputs (user preferences include, for example, advertisement preferences for natural language responses, paragraph [0089]), the at least one first memory encoded with first instructions that, when executed, cause the first processor to: receive at least one input indicative of a natural-language text string (a prompt 710 is received from a user 702, paragraph [0103]); and
provide the natural-language text string as a natural-language text prompt to a chat application operating on the user device (see Fig. 5, user prompts provided in a chat application, paragraph [0092]); and
a remote device communicatively connected to the user device (a server, paragraph [0085]), the remote device comprising:
a second processor (general purpose processor, paragraph [0228]); and
at least one second memory (RAM, paragraph [0229]) encoded with second instructions that, when executed, cause the second processor to:
receive the natural language text prompt from the user device (a prompt 710 is received from a user 702, paragraph [0103]);
receive the at least one user preference from the user device (a user profile comprising user preferences is provided from a user’s device, paragraphs [0091] and [0103]);
receive at least one first operator preference indicative of at least one second characteristic, preferred by an operator of the server, of the natural-language outputs (brand/advertiser preferences 716 are received, paragraph [0103]);
modify a prompt for the machine-learning language model based on the received at least one user preference and the at least one first operator preference to generate a modified system prompt (ad module 750 modifies prompt 710 based on user preferences 714 and brand/advertiser preferences 716, paragraph [0103]);
provide the modified prompt as input to the machine-learning language model to generate a natural-language text output (updated prompts 720 are provided to a generative language model system 704, paragraph [0107]);
transmit the natural-language text output to the user device (output 722 is provided to the user device, see Fig. 7 and paragraph [0107])).
While Taveira discloses modifying a natural language prompt according to user preferences and operator preferences, Taveira does not expressly disclose first providing a modified system prompt, then subsequently providing the natural-language text prompt provided by the user as a separate prompt.
Deutsch discloses a method for providing custom model instructions based on user preferences, comprising providing, by a server, a modified system prompt as an initial input to a machine-learning language model (in response to receiving a natural language prompt from a user, paragraph [0061]; custom instructions related to a user’s preferences are provided to a natural language model as a system prompt, paragraphs [0067] and [0071-0072]); and
providing, by the server and after providing the modified system prompt, the natural-language text prompt as an input to the machine-learning language model to generate a natural-language text output (the natural language prompt from the user is also provided to the natural language model to generate a natural language response, paragraphs [0067] and [0071-0072]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to split the prompt of Taveira into an initial system prompt, followed by the natural language text prompt, because providing custom model instructions as a system prompt prefix before a user input improves the accuracy, efficiency, trainability, and generation of language model responses based on custom instructions, as taught by Deutsch (paragraphs [0005-0006]).
Allowable Subject Matter
Claims 6-19 are 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:
Claim 6 requires the at least one second preferred characteristic comprises an operator-preferred vendor and the at least one third preferred characteristic comprises an operator-preferred data source for context injection. While Taveira discloses the second preferred characteristic comprises an operator-preferred vendor (brand preference, paragraph [0103]), Taveira does not disclose or suggest the third preferred characteristic comprises an operator-preferred data source for context injection. Additionally, Deutsch and the additional prior art of record do not disclose or suggest this recited combination of a second preferred characteristic comprises an operator-preferred vendor and a third preferred characteristic comprises an operator-preferred data source for context injection.
Claims 7-19 would be allowable for their dependence on claim 6.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Afram et al., Zawideh et al., De Wynter et al., Mesard et al., Lin et al., Ryskamp, and Liu et al. disclose additional methods/systems for modifying prompts based on user and/or operator preferences.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN LOUIS ALBERTALLI whose telephone number is (571)272-7616. The examiner can normally be reached M-F 8AM-3PM, 4PM-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, Bhavesh Mehta can be reached at 571-272-7453. 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.
BLA 7/20/26
/BRIAN L ALBERTALLI/ Primary Examiner, Art Unit 2656
Prosecution Insights