DETAILED ACTION
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office 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 .
Request for Continued Examination
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 01/20/2026 has been entered.
Response to Amendments and Arguments
Regarding a rejection to claims under 35 U.S.C. §101, applicant amended independent claims by adding additional limitations. Applicant presented arguments (Remarks, pages 7-10) by citing USPTO’s Appeals Review Panel (ARP) recent decisions (Remarks, page 8). In particular, applicant stated (Remarks, page 9) that “Therefore, Applicant respectfully submits that the Specification clearly articulates an improvement to the technologies of responding to user inquiries using voice assistants. Further, Applicant submits that this improvement is properly reflected in the claims, for example the elements relating to generating the input prompt by combining the user input, the sample prompt, and the additional information.”
By reviewing the amended claims and applicant’s arguments, the examiner decided to withdrawn the rejection under §101.
Regarding a rejection to claims under 35 U.S.C. §103, applicant amended independent claims. Applicant argued that previously cited prior art references fail to teach certain newly added limitations recited in independent claims (Remarks pages 11-14).
The examiner performed an update search and discovered a relevant reference to Li et al. (US PG Pub. 2022/0229832, referred to as Li). Li discloses an iterative procedure to prompt a GPT3 model to generate a desired content. The prompt design component (Fig. 1, #115) combines various information to generate a prompt as an input to the GPT3 model (Li, [0021-0025], prompting a GPT-3 transformer model using various information including a user’s query, a prompt with a selected topic, hints, suggestions as well as a blocklist to prevent generating a bias content, Fig. 1 and Fig. 2).
In the following rejection, the examiner combines the Li reference with the previously cited primary reference (Jolley, US PG Pub. 2017/0242886) to reject the amended claims. Applicant arguments (Remarks, pages 11-14) are moot because the arguments do not apply to a combined teaching of Jolley in view of Li being used in the following rejection.
Claim Rejections - 35 USC § 112
Claims 1-6 and 8-15 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor, or for pre-AIA the applicant regards as the invention.
Independent claims 1 and 11 recites a term “the prompt”. Since antecedent limitations mention “a sample prompt” and “a plurality of sample prompts”. It is unclear whether a claimed “the prompt” refers a previously mentioned “a sample prompt” or a different prompt among the recited “a plurality of sample prompts”. If “the prompt” refers to previously mentioned “a sample prompt”, a claim must use term consistently. Dependent claims include all limitations of their corresponding independent claims. All dependent claims are rejected.
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 of this title, 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 1-3 and 10-13 are rejected under 35 U.S.C. §103 as being unpatentable over Jolley et al. (US PG Pub. 2017/0242886, referred to as Jolley) in view of Li et al. (US PG Pub. 2022/0229832, referred to as Li).
Jolley discloses a user could interact with an intelligent agent (i.e., a chatbot) using either writing text or using voice ([0080-0082], [0092], Fig. 5A / 5B). The intelligent agent system determines additional information such as user intents, domains, user’s location, time of day ([0059], [0080-0083], [0207-0210], Fig. 8B). The system generates queries based on user’s requests, user’s intents and domains for searching a knowledge base ([0032], [0034], [0221-0225]). The responses based on searching results are presented to the user ([0074-0076], Fig. 5B).
Li discloses an iteratively procedure to create a prompt as an input to a GPT3 model for generating desired contents. Li discloses a prompt is created by including various information, e.g., a user’ query, intent, topic, hints and suggestions in order to obtain wanted results (Li, [0020-0021], Fig. 1). In particular, Li further discloses a created prompt may include formatting information (Li, [0027], “making the background purple”, [0038], slides formatting request).
Regarding claims 1 and 11, Jolley discloses an apparatus and a method configured to communicate with a server including a large language model, (Jolley, [0040-0041], [0207-0210], Fig. 2 and Fig. 5A, a computer implemented intelligent agent having interactions with a user by providing responses based on user’s input and additional information such as user’s intent, domain; [0080], [0049], [0443], The intelligent agent could understanding user’s natural language queries and implies having a large language model) the method comprising:
obtain user input acquired from a user (Jolley, [0092], Fig. 5A, an intelligent receives a user’s question; Fig. 8B, #832);
selecting a sample prompt related to the user input from among the stored plurality of sample prompts (Jolley, [0032], [0117], [0439], Fig. 8B, #836, parsing user’s question against a set of interpretations; [0127-0129], various question types; Fig. 5B shows an initial search query for vegan soup recipe, and then further shows vegan restaurant based on user’s intent);
obtaining additional information corresponding to the user input (Jolley, [0038], [0049], [0068], [0208-0209], Fig. 8B, #838, determining user’s intent and question domains, e.g., “movie” or “food”; [0273], user’s location, [0354], time of day; Examiner notes, acquired user’s intents, domains, user location, time of day corresponds to a claimed “additional information”);
obtaining an input prompt based on the user input, the sample prompt, and the additional information (Jolley, [0029], [0033-0034], [0092], [0205], [0272], Fig. 5B, providing a search query to a search engine based on user’s question, intent and domain and obtaining search results);
transmitting the input prompt to the server (Jolley, [0048], [0064], providing search queries to a remote search engine); and
providing the response information to the user (Jolley, [0273], [0278], Fig. 5B, presenting search results received from a search engine).
Jolley discloses a user could interact with an intelligent agent (i.e., a chatbot) using natural languages (Jolley, [0080-0082], [0092], Fig. 5A / 5B). Jolley further discloses queries based on user’s requests, user’s intents (i.e., “additional information”) and domains for searching a knowledge base (Jolley, [0032], [0034], [0221-0225]). The responses based on searching results are presented to the user (Jolley, [0074-0076], Fig. 5B). Jolley does not disclose newly added limitations. However, Li discloses features defined by the newly added limitations.
Li discloses generating wanted contents by prompting a GPT3 model using a prompt design component (Li, [0015], [0021-0023]). Li discloses a prompt design component creating a prompt for the GPT3 model by combining various information such as queries from a user, a sample prompt of a selected topic, prompt library and examples, suggestions, hints (Li, [0020-0023], [0030], [0033], Fig. 4).
Both Jolley and Li are related to obtaining information by providing questions / queries. It would have been obvious to a person having ordinary skill in the art at the time the invention was filed to combine Jolley’s teaching with Li’s teaching to create prompts for a language model to generate wanted contents by combining various information to the GPT3 model. One having ordinary skill in the art would have been motivated to make such a modification so that the system can output user’s wanted outputs and filter out unwanted contents (Li, [0021]).
Regarding claims 2 and 12, Jolley in view of Li further discloses the sample prompt comprises at least one of description information representing the sample prompt, at least one example information, and response structure information (Jolley, Fig. 5B, “I know some vegan soup recipes”).
Regarding claims 3 and 13, Jolley in view of Li further discloses identifying a task corresponding to the user input, and selecting the sample prompt based on the identified task (Jolley, [0207-0210], different interpretation classes such as movie, food, airport).
Regarding claim 10, Jolley in view of Li further discloses the user input comprises at least ONE of audio information, text information, and image information (Jolley, [0092], user could utter or write “where can I get a pizza”).
Claims 4-6, 8-9 and 14-15 are rejected under 35 U.S.C. 103 as being unpatentable over Jolley in view of Li and further in view of Hwang (US PG Pub. 2021/0065685, referred to as Hwang).
Jolley discloses a user could interact with an intelligent agent using natural language (Jolley, Fig. 5A-5D). Jolley discloses the intelligent agent determines user’s intent based on additional information including context information (Jolley, [0197-0203]), user’s profile, locations or dialog histories (Jolley, [0059], [0199], [0204]). Although many features in these dependent claims are implicitly disclosed or obvious modification, the examiner combines Jolley with Hwang, which is a published patent application by the same assignee (Samsung).
Hwang discloses a user could have conversations with a virtual personal assistant to request information (Hwang, Fig. 1A). The virtual personal assistant in Hwang reference is similar to the voice assistant of the instant application.
Regarding claims 4 and 14, Jolley in view of Li discloses an intelligent agent response to user’s questions by considering user’s intent, context such as location, user’s historical statements. Jolley further shows the intelligent agent communicates with remote servers on the internet (Jolley, Fig. 2). Hwang also shows receiving additional information from a server (Hwang, [0100-0104], [0106-0107], [0240-0243], Fig. 8).
Jolley in view of Li and Hwang are dealing with using an intelligent agent / voice agent. It would have been obvious to a person having ordinary skill in the art at the time the invention was filed to combine Jolley teaching with Hwang’s teaching to request another server to send additional information such as context, location information. One having ordinary skill in the art would have been motivated to make such a modification because the information may be store on different computers and could be accessed through network. all the claimed elements were known in the prior art and one skilled in the art could have combined the elements as claimed by known methods, and in the combination each element merely would have performed the same function as it did separately. “A combination of familiar elements according to known methods is likely to be obvious when it does no more than yield predictable results.” KSR, 550 U.S. ___, 82 USPQ2d at 1395 (2007). One of ordinary skill in the art would have recognized that the results of the combination were predictable.
Regarding claims 5 and 15, Jolley in view of Li and Hwang further discloses acquiring a first probability of generating a complete response based on the user input, and acquiring the additional information based on the first probability being less than a threshold probability (Jolley, [0112], [0195], Fig. 5B, the initial returned search result is not what the user’s intend result (being less than a threshold probability), Hwang, [0059], [0064], [0075-0078], [0135-0140], Fig. 8, S816).
Regarding claim 6, Jolley in view of Li and Hwang further discloses acquiring a second probability of generating the complete response based on the user input and at least one piece of sample additional information from among a plurality of pieces of sample additional information, and acquire the additional information based on at least one piece of sample additional information based on the second probability being greater than or equal to the threshold probability (Jolley, Fig. 5B, after including additional information (restaurant domain), the returned results corresponds to user’s intended search results; Hwang, [0058-0059], [0064-0065], [0071-0073], [00276-00277]).
Regarding claim 8, Jolley in view of Li and Hwang further discloses acquiring a user log corresponding to user activity within a threshold time, and select the sample prompt based on the user log and the user input (Jolley, [0035], [0052], [0059], considering previous conversation and historical statements; Hwang, [0290]).
Regarding claim 9, Jolley in view of Li and Hwang further discloses acquiring a user log corresponding to user activity within a threshold time, and acquire the additional information based on the user log and the user input (Jolley, [0035], [0052], [0059], [0199], [0204], determining user’s intent and context by considering history of prior queries and historical statements; Hwang, [0053], [0142]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JIALONG HE whose telephone number is (571)270-5359. The examiner can normally be reached on Monday-Thursday, 7:00AM-4:30PM, ALT. Fridays, EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Pierre Desir can be reached on (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 an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JIALONG HE/Primary Examiner, Art Unit 2659