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 .
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) 1-3,13-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang U.S. PAP2023/0018489 A1 in view of Khani U.S. PAP 2024/0370662 A1.
Regarding claim 1 Jiang teaches a training method for a machine learning model and executed by an electronic device (acquiring a structured QA model according to an embodiment of the present disclosure. The method is performed by an apparatus for acquiring a structured QA model, see par. [0021]), the training method comprising:
obtaining a plurality of training samples ( training samples corresponding to N types of structured QA database are acquired, see par. [0022]);
inputting each of the training samples to a first machine learning model to obtain a feature vector correspondingly (The encoder encodes the sample sequence to obtain vector representations of the Tokens in the sample sequence., see par. [0039]);
and training a second machine learning model according to the representative training samples (training a text generation model by using the training samples to obtain the structured QA model, see par. [0036]).
However Jiang does not teach clustering the feature vectors corresponding to the training samples to obtain a plurality of groups, wherein each of the groups includes a portion of the feature vectors; extracting a representative feature vector from each of the groups, wherein the representative feature vector corresponds to a representative training sample among the training samples, and a quantity of the representative training samples corresponding to the groups is less than a quantity of the training samples.
In a similar field of endeavor Khani teaches a system and method and for method for optimizing performance of a natural language processing (NLP) model, see abstract. Khani teaches The clustering engine 210 implements agnostic clustering of the validation dataset 240 by using general purpose embeddings, such as embeddings extracted from sentence bidirectional encoder representations from transformers (BERT) that are implemented in sentence transformers, see par. [0029] (clustering the feature vectors corresponding to the training samples to obtain a plurality of groups, wherein each of the groups includes a portion of the feature vectors). [0097] generating a prompt for submission as an input to a large language model (LLM) to prompt the LLM to automatically generate synthetic training data for the identified cluster; [0098] providing the prompt to the LLM; [0099] receiving from the LLM the synthetic training data (extracting a representative feature vector from each of the groups, wherein the representative feature vector corresponds to a representative training sample among the training samples, and a quantity of the representative training samples corresponding to the groups is less than a quantity of the training samples); and [0100] using the synthetic training data to further train the NLP model to improve the performance of the NLP model with respect to the identified cluster.
It would have been obvious to one of ordinary skill in the art to combine the Jiang invention with the teachings of Khani for the benefit of improve the performance of the NLP model with respect to the identified cluster, see par. [0100].
Regarding claim 2 Jiang teaches the training method of claim 1, wherein the second machine learning model is a pretrained model (The Embedding may be performed by the encoder or a pre-trained model independent of the encoder, see par. [0041]).
Regarding claim 3 Khani teaches the training method of claim 1, wherein the step of clustering the feature vectors corresponding to the training samples to obtain the groups includes:
calculating similarities between the feature vectors; and if the similarity between two of the feature vectors is greater than a similarity threshold, clustering the two of the feature vectors into same one of the groups (The targeted data generation system 110 may use the LLM 130 to generate training data by prompting the LLM 130 to create similar in-cluster examples for the identified data clusters, see par. [0024]).
Regarding claim 13 Jiang teaches an electronic device, including: a memory, storing a plurality of instructions; a processor, communicatively connected to the memory, (an electronic device is provided, including at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform a method for acquiring a structured question-answering (QA) model, see par. [0007]) and configured to execute the instructions to perform a plurality of steps:
obtaining a plurality of training samples ( training samples corresponding to N types of structured QA database are acquired, see par. [0022]);
inputting each of the training samples to a first machine learning model to obtain a feature vector correspondingly (The encoder encodes the sample sequence to obtain vector representations of the Tokens in the sample sequence., see par. [0039]);
and training a second machine learning model according to the representative training samples (training a text generation model by using the training samples to obtain the structured QA model, see par. [0036]).
However Jiang does not teach clustering the feature vectors corresponding to the training samples to obtain a plurality of groups, wherein each of the groups includes a portion of the feature vectors; extracting a representative feature vector from each of the groups, wherein the representative feature vector corresponds to a representative training sample among the training samples, and a quantity of the representative training samples corresponding to the groups is less than a quantity of the training samples.
In a similar field of endeavor Khani teaches a system and method and for method for optimizing performance of a natural language processing (NLP) model, see abstract. Khani teaches The clustering engine 210 implements agnostic clustering of the validation dataset 240 by using general purpose embeddings, such as embeddings extracted from sentence bidirectional encoder representations from transformers (BERT) that are implemented in sentence transformers, see par. [0029] (clustering the feature vectors corresponding to the training samples to obtain a plurality of groups, wherein each of the groups includes a portion of the feature vectors). [0097] generating a prompt for submission as an input to a large language model (LLM) to prompt the LLM to automatically generate synthetic training data for the identified cluster; [0098] providing the prompt to the LLM; [0099] receiving from the LLM the synthetic training data (extracting a representative feature vector from each of the groups, wherein the representative feature vector corresponds to a representative training sample among the training samples, and a quantity of the representative training samples corresponding to the groups is less than a quantity of the training samples); and [0100] using the synthetic training data to further train the NLP model to improve the performance of the NLP model with respect to the identified cluster.
It would have been obvious to one of ordinary skill in the art to combine the Jiang invention with the teachings of Khani for the benefit of improve the performance of the NLP model with respect to the identified cluster, see par. [0100].
Regarding claim 14 Jiang teaches the electronic device of claim 13, wherein the second machine learning model is a pretrained model (The Embedding may be performed by the encoder or a pre-trained model independent of the encoder, see par. [0041])..
Regarding claim 15 Khani teaches the electronic device of claim 13, wherein the step of clustering the feature vectors corresponding to the training samples to obtain the groups includes: calculating similarities between the feature vectors; and if the similarity between two of the feature vectors is greater than a similarity threshold, clustering the two of the feature vectors into same one of the groups (The targeted data generation system 110 may use the LLM 130 to generate training data by prompting the LLM 130 to create similar in-cluster examples for the identified data clusters, see par. [0024]).
Claim(s) 4, 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang U.S. PAP2023/0018489 A1 in view of Khani U.S. PAP 2024/0370662 A1 further in view of Jamei U.S. PAP 2025/0355921A1.
Regarding claim 4 Jiang in view of Khani does not teach the training method of claim 3, wherein the step of extracting the representative feature vector from each of the groups includes: establishing a graph for a first group among the groups, wherein the graph comprises a plurality of vertices and at least one edge, the vertices correspond to the feature vectors in the first group, and the at least one edge indicates that the similarity between the feature vectors in the first group is greater than the similarity threshold; and setting one of the vertices with a largest number of connections as the representative feature vector.
In the same field of endeavor Jamei teaches generating a graph based on the executed search, the graph comprising a set of nodes representing a topic or variable and edges connecting a first node to a second node or a first node to multiple nodes, wherein each edge represents one of the statistical or mechanistic relationships extracted from the source materials; clustering or grouping the nodes and summarizing data or information represented by nodes or edges contained in each cluster or group, wherein each cluster or group includes a set of nodes representing semantically similar results of the search, and wherein the clustering, grouping, or summarizing is performed at least in part using expert guidance, expert provided rules, or expert provided conditions, see par. [0315-0316].
It would have been obvious to one of ordinary skill in the art to combine the Jiang in view of Khani invention ith the teachings of Jamei for the benefit of synthesizing or enhancing the clustered or grouped nodes using retrieval augmented generation, see par. [0317].
Regarding claim 16 Jiang in view of Khani does not teach the electronic device of claim 15, wherein the step of extracting the representative feature vector from each of the groups includes: establishing a graph for a first group among the groups, wherein the graph comprises a plurality of vertices and at least one edge, the vertices correspond to the feature vectors in the first group, the at least one edge indicates that the similarity between the feature vectors in the first group is greater than the similarity threshold; and setting one of the vertices with a largest number of connections as the representative feature vector.
In the same field of endeavor Jamei teaches generating a graph based on the executed search, the graph comprising a set of nodes representing a topic or variable and edges connecting a first node to a second node or a first node to multiple nodes, wherein each edge represents one of the statistical or mechanistic relationships extracted from the source materials; clustering or grouping the nodes and summarizing data or information represented by nodes or edges contained in each cluster or group, wherein each cluster or group includes a set of nodes representing semantically similar results of the search, and wherein the clustering, grouping, or summarizing is performed at least in part using expert guidance, expert provided rules, or expert provided conditions, see par. [0315-0316].
It would have been obvious to one of ordinary skill in the art to combine the Jiang in view of Khani invention ith the teachings of Jamei for the benefit of synthesizing or enhancing the clustered or grouped nodes using retrieval augmented generation, see par. [0317].
Claim(s) 8-12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang U.S. PAP2023/0018489 A1 in view of Jamei U.S. PAP 2025/0355921A1.
Regarding claim 8 Jiang teaches a refinement method executed by an electronic device, the refinement method comprising:
(a) obtaining an original sample (Acquire training samples corresponding to N types of structured QA database, see par. [0007]).
However Jiang does not teach (b) querying an external database according to the original sample to obtain supplementary data; (c) inputting the original sample and a first prompt to a first machine learning model to obtain review data; and (d) inputting the original sample, the supplementary data, the review data, and a second prompt to a second machine learning model to obtain a refined sample corresponding to the original sample, wherein the second machine learning model is different from the first machine learning model.
In the same field of endeavor Jamei teaches (b) querying an external database according to the original sample to obtain supplementary data (databases are accessed, see par. [0090]);
(c) inputting the original sample and a first prompt to a first machine learning model to obtain review data ( Search, which conducts a search in the sources and extracted findings in response to a user query, see par. [0094]); and
(d) inputting the original sample, the supplementary data, the review data, and a second prompt to a second machine learning model to obtain a refined sample corresponding to the original sample, wherein the second machine learning model is different from the first machine learning model (FIG.1(o) ; Converting the retrieved information to readable text may be performed using the following steps. The summarized clusters produced from retrieved information based on a user's query (as described in previous steps) are created. The clusters are then grouped under headings for the purpose of organizing and presenting the results in more easily readable paragraphs. These headings are created from LLMs via directed prompt, and each of the cluster summaries are “assigned” to a heading. An LLM then generates a paragraph based on a prompt that instructs the LLM to prepare text based on the heading and the assembled summaries., see par. [0204]).
Regarding claim 9 Jamei teaches the refinement method of claim 8, further comprising: replacing the original sample with the refined sample and repeatedly executing the step (c) and the step (d) (raining uses general-purpose methods to iteratively determine the weights for intermediate and final feature neurons, see par. [0359]).
Regarding claim 10 Jiang teaches the refinement method of claim 8, wherein the original sample includes a question and an answer, and the step (b) comprises: querying the external database according to the question to obtain the supplementary data (acquiring a structured QA model according to an embodiment of the present disclosure. The method is performed by an apparatus for acquiring a structured QA model, see par. [0021]).
Regarding claim 11 Jamei teaches the refinement method of claim 10, wherein the original sample comprises a text, the first machine learning model is a language model, and the first prompt is configured to instruct evaluating correctness, fluency, and completeness of the answer (An example of the Search algorithm is illustrated in FIG. 1(l). As shown in the figure, the search algorithm takes user queries (entered into a standard text box), and processes them to correct for possible spelling errors or to perform disambiguation of terms (as non-limiting examples) and submits them to Pubmed's Entrez search API, see par. [0175]).
Regarding claim 12 Jamei teaches the refinement method of claim 10, wherein the second machine learning model is a language model, and the second prompt is configured to instruct adjusting the answer according to the supplementary data and the review data (LLM takes expansion prompt (each heading and its assigned clusters are presented to the LLM for content expansion. These expansion requests are executed in parallel to enhance efficiency. The expansion prompt contains specific instructions for the LLM to adhere strictly to the provided context, barring the addition of term definitions or acronym explanation, see par. [0209].
Allowable Subject Matter
Claims 5-7 and 17-19 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.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Xing ‘977 teaches visual clustering aims to train a clustering model based on visual embeddings from label-annotated datasets. Visual clustering may include first constructing a k-NN graph using the features (e.g., embedding vectors), using a GNN to encode the information from this graph, and decoding predicted graph node attributes or edges into cluster assignments.
Liu ‘181 teaches a tuning stage can be performed on each of the task-specific layers in sequence again using further batches of appropriate training data. This process can continue over several iterations until the tuning stage is complete, see par. [0060].
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael Ortiz-Sanchez whose telephone number is (571)270-3711. The examiner can normally be reached Monday- Friday 9AM-6PM.
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.
/MICHAEL ORTIZ-SANCHEZ/Primary Examiner, Art Unit 2656
Prosecution Insights