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 .
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 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.
Priority
Acknowledgment is made of applicant's claim for foreign priority based on Chinese application CN202410141187.0 filed on 01/31/2024. Certified copy of said foreign application has been received.
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) NOVELTY; PRIOR ART.—A person shall be entitled to a patent unless—
(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention; or
(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.
PNG
media_image1.png
18
19
media_image1.png
Greyscale
(b) EXCEPTIONS.—
(1) DISCLOSURES MADE 1 YEAR OR LESS BEFORE THE EFFECTIVE FILING DATE OF THE CLAIMED INVENTION.—A disclosure made 1 year or less before the effective filing date of a claimed invention shall not be prior art to the claimed invention under subsection (a)(1) if—
(A) the disclosure was made by the inventor or joint inventor or by another who obtained the subject matter disclosed directly or indirectly from the inventor or a joint inventor; or
(B) the subject matter disclosed had, before such disclosure, been publicly disclosed by the inventor or a joint inventor or another who obtained the subject matter disclosed directly or indirectly from the inventor or a joint inventor.
(2) DISCLOSURES APPEARING IN APPLICATIONS AND PATENTS.—A disclosure shall not be prior art to a claimed invention under subsection (a)(2) if—
(A) the subject matter disclosed was obtained directly or indirectly from the inventor or a joint inventor;
(B) the subject matter disclosed had, before such subject matter was effectively filed under subsection (a)(2), been publicly disclosed by the inventor or a joint inventor or another who obtained the subject matter disclosed directly or indirectly from the inventor or a joint inventor; or
(C) the subject matter disclosed and the claimed invention, not later than the effective filing date of the claimed invention, were owned by the same person or subject to an obligation of assignment to the same person.
Claims 1-2, 7, 10-11, 16, and 19-20 are rejected under 35 USC 102(a)(2) as being anticipated by Zhou et al. (CN 116842126B, see attached IP.com translation).
Regarding Claims 1, 10, and 19, Zhou discloses an electronic device (p. 8, “The third aspect of the invention provides a system for realizing accurate output of knowledge base by LLM”), comprising:
a memory, configured to store a computer program (p. 8, “A second aspect of the present invention provides a computer-readable storage medium in which program instructions are stored”); and
one or more processors, configured to, when the computer program is executed (p. 8, “A second aspect of the present invention provides a computer-readable storage medium in which program instructions are stored for performing the above-described method for realizing accurate output of a knowledge base using LLM when the program instructions are run”; i.e., a computer system running / performing program instruction inherently has a processor for so running / performing), perform:
in response to obtaining of a target input message (p. 9, s20, “acquiring a problem of a user and carrying out vectorization processing to obtain a problem vector”), outputting a first response message based on a target knowledge base (p. 9, s30, “matching the problem vector with a knowledge vector data set to obtain M knowledge vectors with highest matching degree”, the knowledge vector data set was obtained per s10 by carrying out vectorization processing on knowledge in a knowledge base to obtain a knowledge vector data set containing a plurality of knowledge vectors); and
inputting the first response message into at least one target large language model, and outputting a second response message (p. 9, s40, “performing text processing on the M obtained knowledge vectors to obtain a corresponding problem text serving as a prompt” and s50, “submitting the obtained prompt to N LLM models to obtain N output texts”), wherein the at least one target large language model is different from the target knowledge base (p. 9, s50, N LLM models; per p. 14, 3. LLM prediction “Commonly used LLMs include GPT, BERT, etc.”; compare p. 10, 1. Construction of knowledge base; “…using the existing knowledge sources such as encyclopedia, dictionary, professional book, manually writing knowledge points”); and
a matching degree between the second response message and the target input message (p. 9, s61 “carrying out vectorization processing on the obtained N output texts to obtain N output vectors”, s62 “carrying out correlation analysis on each output vector and a knowledge base to obtain the correlation of each output vector”; per p. 14, 2. Prompt generation, “…For example, “question:” may be added as a question description before each predicted text, and “answer:” may be added as an answer prompt at the end” (
PNG
media_image2.png
49
685
media_image2.png
Greyscale
) and “assume that there are N LLM models,…N different predicted texts can be obtained” (
PNG
media_image3.png
26
586
media_image3.png
Greyscale
); i.e., prompt requires the N output vectors to comprise question description (target input message) before respective N predicted text (second response message) such that correlation analysis correlates question description (target input message) and corresponding N answers (second response message) with a knowledge base (comprising M knowledge vectors with highest matching degree with the problem vector per p. 9, step s30) to calculate respective correlation degree between the target input message and the second response message) is higher than (p. 8, 1. Improving the accuracy of knowledge retrieval, “according to the invention, through vectorization expression of questions and knowledge, preliminary matching is performed by calculating the similarity between vectors” and 2. Enhancing the correctness and fluency of knowledge expression, “By submitting the matched knowledge vector to LLM to generate response text as a prompt…compared with directly outputting the retrieved knowledge text, the response text synthesized by the method is more smooth and accurate in grammar and semanteme”; i.e., compare to the preliminary matching of problem vector with M obtained knowledge vectors in the knowledge base, the correlations of the N output vectors comprising the problem description / target input message and the N answers (second response messages) with M knowledge vectors in the knowledge base are more smooth and accurate) a matching degree between the first response message and the target input message (p. 12, “The purpose of step S30 is to match the knowledge vector data set with the problem vector, so as to obtain M knowledge vectors with the highest matching degree…measured the degree of semantic relatedness of the question to each knowledge point”).
Further regarding claim 19, Zhou discloses a non-transitory computer-readable storage medium containing a computer program that when being executed, causes one or more processors to perform the method of claim 1 and device function of claim 10 (p. 8, “A second aspect of the present invention provides a computer-readable storage medium in which program instructions are stored for performing the above-described method for realizing accurate output of a knowledge base using LLM when the program instructions are run”).
Regarding Claims 2, 11, and 20, Zhou discloses wherein for outputting the first response message based on the target knowledge base in response to the obtaining of the target input message (p. 12, “Similar to vectorizing the knowledge points in step S10, the step S20 first entails vectorizing the input natural language question, expressed as a dense vector of fixed dimensions q. Specifically, the text vectorization model introduced in step S10, such as Word2Vec, BERT, etc., may be used to encode the question text to obtain the question vector”), the one or more processors are configured to perform at least one of following:
in response to the obtaining of the target input message, extracting a keyword from the target input message, and inputting extracted keyword into a local knowledge base of an electronic device and/or a first large language model to output the first response message;
in response to the obtaining of the target input message, after an enhancement processing is performed on the target input message (p. 11, “If the knowledge points are expressed by natural language, the text can be regarded as a sequence, each word is mapped into a dense vector by a word embedding method, and the vector representation of the whole text sequence is obtained through a model…These models can learn semantic features of text, mapping the text into a semantic vector space of fixed dimensions”; i.e., enhance the input natural language question by mapping question text into semantic vector space of fixed dimensions), inputting enhanced target input message into the local knowledge base of the electronic device and/or the first large language model to output the first response message (p. 12, “the purpose of step S30 is to match the knowledge vector data set with the problem vector, so as to obtain M knowledge vectors with the highest matching degree”); or
in response to the obtaining of the target input message, determining to output the first response message using the local knowledge base of the electronic device and/or the first large language model based on an attribute message and/or content of the target input message, wherein the first large language model is trained based on a first knowledge base different from the local knowledge base.
Regarding Claims 7 and 16, Zhou discloses wherein for inputting the first response message into the at least one target large language model and outputting the second response message (p. 9, s40, “performing text processing on the M obtained knowledge vectors to obtain a corresponding problem text serving as a prompt” and s50, “submitting the obtained prompt to N LLM models to obtain N output texts”), the one or more processors are configured to:
input the first response message to a plurality of target large language models respectively to generate a plurality of response messages (p. 9, s50, submitting the obtained prompt to N LLM models to obtain N output texts; p. 14, “assume that there are N LLM models,…N different predicted texts can be obtained”); and
compare similarities between the plurality of response messages outputted by the plurality of target large language models and output a response message with a highest similarity as the second response message (p. 9, s60, “performing correlation analysis on the obtained N output texts, and taking the output text with the highest correlation as an output result”, s62, “carrying out correlation analysis on each output vector and a knowledge base to obtain the correlation of each output vector”, s63, “if the correlation degree of the output vector with the largest correlation degree is larger than a correlation degree threshold value, taking the output text corresponding to the output vector with the largest correlation degree as an output result”).
Claim Rejections - 35 USC § 103
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 103 that form the basis for the rejections under this section made 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.
Claims 3 and 12 are rejected under 35 USC 103(a) as being unpatentable over Zhou et al. (CN 116842126B, see attached IP.com translation) in view of Huang et al. (CN117216210A).
Regarding Claims 3 and 12, Zhou discloses wherein for inputting the enhanced target input message into the local knowledge base of the electronic device and/or the first large language model to output the first response message after the enhancement processing is performed on the target input message (p. 11, “If the knowledge points are expressed by natural language, the text can be regarded as a sequence, each word is mapped into a dense vector by a word embedding method, and the vector representation of the whole text sequence is obtained through a model…These models can learn semantic features of text, mapping the text into a semantic vector space of fixed dimensions”; i.e., enhance the input natural language question by mapping question text into semantic vector space of fixed dimensions; p. 12, “the purpose of step S30 is to match the knowledge vector data set with the problem vector, so as to obtain M knowledge vectors with the highest matching degree”).
Zhou does not teach the one or more processors are configured to preform at least one of following:
performing a keyword expansion processing after extracting the keyword from the target input message and/or performing an intent expansion processing on the target input message, and inputting message data after expansion processing into the local knowledge base of the electronic device or the first large language model to output the first response message;
after extracting the keyword from the target input message, obtaining a response message matching the keyword from the local knowledge base of the electronic device; and in response to that the response message matching the keyword is not obtained, after performing the keyword expansion processing, inputting a keyword expansion processing result into the local knowledge base or the first large language model to output the first response message; or
after extracting the keyword from the target input message, obtaining the response message matching the keyword from the local knowledge base of the electronic device; and in response to that the response message matching the keyword is not obtained, inputting the target input message after performing the intent expansion processing into the first large language model to output the first response message.
Huang teaches performing a keyword expansion processing after extracting keyword from target input message and/or performing an intent expansion processing on the target input message (p. 8, step 102 “extracting initial keywords in the problem information;…Specifically, the problem information may be input to the BERT model, so that the initial keywords are extracted through the BERT model. The initial keyword is the most important phrase in the question information, and can represent the operation intention of the identification information”; p. 9, Step 103, “vocabulary expansion processing is carried out based on the initial keywords…an expansion word similar to a semantic meaning of the initial keyword may be searched”, and step 104, “performing word sense enhancement processing on the problem information based on the target keywords to obtain a first vector”), and inputting message data after expansion processing into the local knowledge base of the electronic device or the first large language model to output first response message (p. 9, step 105, “performing similarity matching on the first vector and each second vector in a pre-acquired vector library to obtain a first target vector”).
It would’ve been obvious to one ordinarily skilled in the art before the effective filing date of the invention to perform a keyword expansion processing after extracting the keyword from the target input message (compare Huang, p. 8 “Specifically, the problem information may be input to the BERT model, so that the initial keywords are extracted through the BERT model. The initial keyword is the most important phrase in the question information, and can represent the operation intention of the identification information” with Zhou, p. 12, “the text vectorization model introduced in step S10, such as Word2Vec, BERT, etc., may be used to encode the question text to obtain the question vector” ) and/or performing an intent expansion processing on the target input message, and inputting message data after expansion processing into the local knowledge base of the electronic device to output the first response message in order to improve the accuracy of the target input message / problem information (Huang, p. 9, step 103 “in order to improve accuracy of problem information, in this embodiment, an expansion process may be performed on an initial keyword”).
Claims 4-6 and 13-15 are rejected under 35 USC 103(a) as being unpatentable over Zhou et al. (CN 116842126B, see attached IP.com translation) in view of Liu et al. (US 2024/0428044 A1).
Regarding Claims 4 and 13, Zhou discloses wherein for determining to output the first response message using the local knowledge base of the electronic device and/or the first large language model based on the attribute message and/or the content of the target input message (p. 12, “the purpose of step S30 is to match the knowledge vector data set with the problem vector, so as to obtain M knowledge vectors with the highest matching degree”), the one or more processors are configured to perform at least one of following:
determining to output the first response message using the local knowledge base of the electronic device and/or the first large language model based on a format attribute of the target input message and/or intent content characterized by the target input message (p. 12, “Similar to vectorizing the knowledge points in step S10, the step S20 first entails vectorizing the input natural language question, expressed as a dense vector of fixed dimensions q. Specifically, the text vectorization model introduced in step S10, such as Word2Vec, BERT, etc., may be used to encode the question text to obtain the question vector”; per p. 11, “These models can learn semantic features of text, mapping the text into a semantic vector space of fixed dimensions”; i.e., given natural language / text format attribute of the question, encode the question text to obtain question vector in semantic vector space of fixed dimensions (i.e., intent content) for matching the knowledge vector data in the knowledge base);
determining to output the first response message using the local knowledge base of the electronic device and/or the first large language model based on a source attribute of the target input message and/or the intent content characterized by the target input message; or
determining to output the first response message using the local knowledge base of the electronic device and/or the first large language model based on a data volume attribute of the target input message and/or instruction content carried by the target input message.
Zhou does not disclose in response to the obtaining of the target input message, determining to output the first response message using the first large language model based on an attribute message and/or content of the target input message, wherein the first large language model is trained based on a first knowledge base different from the local knowledge base (3rd alternative in claim 2, since claim 4 depends on claim 2).
Liu discloses in response to obtaining of a target input message (Fig. 3, Question 102), determining to output a first response message using a local knowledge base of an electronic (¶28, top-K source documents / passages from a database of source documents) and a first large language model based on an attribute message and/or content of the target input message (¶29, concatenate the top k passages with the input question into a single text string (i.e., attribute of input being text) to form an input to LLM 120 such that the LLM 120 is provided with a comprehensive and informative context to enhance the accuracy of the output answer), wherein the first large language model is trained based on a first knowledge base different from the local knowledge base (¶19, LLM such as Generative Pre-trained Transformer 3 has 175 billion parameters).
It would’ve been obvious to one ordinarily skilled in the art before the effective filing date of the invention to in response to the obtaining of the target input message, determine to output the first response message using the first large language model based on an attribute message and/or content of the target input message in order to provide a comprehensive and informative context to enhance the accuracy of the output answer / first response message (Liu, ¶29).
Regarding Claims 5 and 14, Zhou discloses wherein for outputting the first response message based on the target knowledge base in response to the obtaining of the target input message, the one or more processors are configured to: in response to the obtaining of the target input message, input the target input message into a local knowledge base of an electronic device (p. 9, s30, “matching the problem vector with a knowledge vector data set to obtain M knowledge vectors with highest matching degree”, the knowledge vector data set was obtained per s10 by carrying out vectorization processing on knowledge in a knowledge base to obtain a knowledge vector data set containing a plurality of knowledge vectors).
Zhou does not disclose input the target input message into a first large language model to output the first response message.
Liu discloses in response to the obtaining of the target input message, input the target input message into a local knowledge base of an electronic device (¶28, retriever model 110 may select one or more related source documents 112a-n from a database of source documents given an input question 102 to select multiple / top K source documents / passages) and a first large language model respectively (¶29, concatenate top k passages 112a-112n with the input question 102 into a single text string 116 to form an input to LLM 120); and
output the first response message after target processing is performed on response messages respectively outputted by the local knowledge base and the first large language model (Fig. 3, answer 125), wherein the first large language model is trained based on a first knowledge base different from the local knowledge base (¶19, LLM such as Generative Pre-trained Transformer 3 has 175 billion parameters).
It would’ve been obvious to one ordinarily skilled in the art before the effective filing date of the invention to in response to the obtaining of the target input message, input the target input message into a local knowledge base of an electronic device and a first large language model respectively, and output the first response message after target processing is performed on response messages respectively outputted by the local knowledge base and the first large language model in order to provide a comprehensive and informative context to enhance the accuracy of the output answer / first response message (Liu, ¶29).
Regarding Claims 6 and 15, Zhou as modified by Liu discloses wherein for outputting the first response message after the target processing is performed on the response messages respectively outputted by the local knowledge base and the first large language model (Liu, Fig. 3), the one or more processors are configured to perform at least one of following:
configuring different weights for a third response message outputted by the local knowledge base and a fourth response message outputted by the first large language model, respectively; and performing an integration processing on the third response message and the fourth response message based on corresponding weights to obtain the first response message;
performing a match processing on the third response message outputted by the local knowledge base and the fourth response message outputted by the first large language model; and determining a response message that a matching degree exceeds a first threshold as the first response message;
obtaining an evaluation message of the third response message outputted by the local knowledge base and the fourth response message outputted by the first large language model (Liu, ¶39, LLM 120 determines whether input passages 112a-n are able to provide an answer to the question 102 to generate a list of candidate answers based on initial feedback); and based on the evaluation message, outputting at least a part of the third response message and/or the fourth response message as the first response message (Liu, ¶40, at stage 1, curate a pool of potential answers based on each passage 112a-n and question 102. Upon evaluating each answer 125a-n in the answer pool, adopt a passage selection process to discard passages that yield an unknown output by the LLM 120); or
performing a superimposing processing on the third response message outputted by the local knowledge base and the fourth response message outputted by the first large language model to obtain the first response message.
Claims 8-9 and 17-18 are rejected under 35 USC 103(a) as being unpatentable over Zhou et al. (CN 116842126B, see attached IP.com translation) in view of Jones et al. (US 2025/0148258 A1).
Regarding Claims 8 and 17, Zhou discloses wherein for inputting the first response message into the at least one target large language model (p. 9, s50, submitting the obtained prompt to N LLM models to obtain N output texts; p. 14, “assume that there are N LLM models,…N different predicted texts can be obtained”) comprises identifying an intent message characterized by the target input message (p. 12, “Similar to vectorizing the knowledge points in step S10, the step S20 first entails vectorizing the input natural language question, expressed as a dense vector of fixed dimensions q. Specifically, the text vectorization model introduced in step S10, such as Word2Vec, BERT, etc., may be used to encode the question text to obtain the question vector”; per p. 11, “These models can learn semantic features of text, mapping the text into a semantic vector space of fixed dimensions”; i.e., given natural language / text format attribute of the question, encode the question text to obtain question vector in semantic vector space of fixed dimensions (i.e., intent content) for matching the knowledge vector data in the knowledge base) and/or the first response message (p. 12, “…vectorizing the knowledge points in step S10”).
Zhou does not disclose the one or more processors are configured to perform at least one of following:
determining the at least one target large language model based on the intent message,
identifying domains to which the target input message and the first response message belong, or
obtaining an evaluation message of the first response message.
Jones discloses inputting a first response message into a target large language model (¶84 and ¶88, Fig. 6, output embeddings 612 representing words and/or sentences into a model architecture corresponding to a control LLM 202 or any of the domain specific LLMs 210) to:
identifying an intent message characterized by a target input message (¶85, input embeddings function like a dictionary that helps the LLM understand the meaning of words by placing them in an embedding space; per ¶92, LLM 202 performing natural language processing on a query related to one or more subject areas) and/or the first response message (¶88, output embeddings 612 can be used to generate the output text by mapping the model’s predicted probabilities of each token to a corresponding token in a vocabulary; e.g., ¶100, a respective lookup in a knowledge base for generating a response), determining the at least one target large language model based on the intent message, and inputting the first response message into the at least one target large language model determined, wherein a domain to which the at least one target large language model belongs is same as a domain to which the target input message and/or the first response message belongs (¶93, selecting a domain specific LLMs 210 based on one or more subject areas associated with the query; ¶100, one or more domain specific LLMs perform a respective lookup in a knowledge based on the query; i.e., use the respective lookup as output embedding to generate output / response);
identifying domains to which the target input message and the first response message belong (¶88, and Fig. 6, LLM model architecture accepts input embedding 602 and output embeddings 612 used to generate output text; e.g., ¶100, query being the input embedding 602 and respective lookup in knowledge base as output embeddings 612 for generating the output); and in response to that the domains to which the target input message and the first response message belong are same, inputting the first response message into at least one target large language model with a same domain as the first embedded message (¶100, based on the query (input embeddings / target input message) with associated subject areas (¶92), one or more domain specific language models perform a respective lookup in a knowledge base (i.e., lookup output embeddings (first response message) in the respective subject areas / knowledge base to generate output text / one or more responses); or
obtaining an evaluation message of the first response message (¶88, during an inference phase, use output embeddings 612 to generate output text, the output embeddings 612 includes values representing words and sentences that help the model understand the meaning of words per ¶85); determining the at least one target large language model based on domains to which the evaluation message and the first response message belong (¶100, domain specific LLMs perform lookup in a knowledge base as output embeddings; in view of ¶98, control LLM receives one or more query responses / output embeddings from domain specific LLMs); and inputting the first response message into the at least one target large language model determined (¶98, control LLM receives the query responses to formulate a response by coalescing the one or more query responses).
It would’ve been obvious to one ordinarily skilled in the art before the effective filing date of the invention to determine the at least one target large language model based on the intent message, identify domains to which the target input message and the first response message belong, or obtain an evaluation message of the first response message in order to select one or more domain specific target LLMs to answer a query / target input message based on subject areas / domains associated with the query / target input message and respective domain specific knowledge of respective target LLM (Jones, ¶93; compare Zhou, p. 13, “Step S50 is to submit the generated template to N different LLM models to obtain N output texts”).
Regarding Claims 9 and 18, Zhou as modified by Jones discloses wherein for outputting the second response message, the one or more processors are configured to perform at least one of following:
obtaining an evaluation message of a plurality of response messages generated by a plurality of target large language models (Jones, ¶88, during an inference phase, use output embeddings 612 to generate output text, the output embeddings 612 includes values representing words and sentences that help the model understand the meaning of words per ¶85; ¶98, control LLM receives one or more query responses from the one or more domain specific LLM); and based on the evaluation message, determining at least one of the plurality of response messages as the second response message (Jones, ¶98, use one or more query responses to formulate a response by coalescing the one or more query responses);
obtaining the plurality of response messages generated by the plurality of target large language models in sequence, and outputting a response message generated by a last target large language model as the second response message, wherein the last target large language model has a higher matching degree with the target input message than a previous target large language model; or
based on configuration message of an output module of an electronic device, processing the second response message into target media data for output (Zhou, p. 13, “The purpose of step S40 is to generate text from the M knowledge vectors obtained in step S30, and submit the generated text to LLM as a prompt. Step S50 is to submit the generated template to N different LLM models to obtain N output texts”; p. 14, “Commonly used LLMs include GPT, bert, etc.…As an input sequence for LLM, a reply to a prompt may be generated as an answer to the question”).
Conclusion
Prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
US 2025/0355907 A1 discloses an AI broker receiving an incoming request, evaluating the incoming request to determine routing characteristics, selecting specialized language models based on the routing characteristics, computing performance metrics for the selected specialized language models, routing the incoming request to the selected specialized language models based on the performance metrics to generate a final result by receiving and processing results from the selected specialized language models (Abstract).
US 2023/0074406 A1 discloses performing NLU processing on spoken utterance to generate NLU outputs for one or more first party systems and 3rd party systems to generate assistant outputs that are responsive to the spoken utterance captured (¶42). LLM engines can process the assistant outputs to generate LLM outputs (¶43) corresponding to synthesized speech for audible presentation to the user (¶44).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to examiner Richard Z. Zhu whose telephone number is 571-270-1587 or examiner’s supervisor Hai Phan whose telephone number is 571-272-6338. Examiner Richard Zhu can normally be reached on M-Th, 0730:1700.
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.
/RICHARD Z ZHU/Primary Examiner, Art Unit 2654 09/12/2026
Prosecution Insights