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 .
Response to Arguments
Applicant's arguments filed 2/19/26 have been fully considered but they are not persuasive.
Regarding the 35 U.S.C. 103 rejection of claims 1 and 12 with references deLevie and Saxena, Applicant argues that the cited portion of deLevie describes text data being divided into subsets but does not disclose new language “accessing a dataset including a plurality of data subsets, each of the data subsets corresponding to a feature of the dataset features of the dataset corresponding to various data types” (Arguments, pg. 10, fourth para. – pg. 11, first para.).
Examiner respectfully disagrees as deLevie discloses receiving a body of text including documents, as well as a request to divide the text into one or more portions/chunks/pages to be provided as input to a large language model (para. [0025]; para. [0061]; para. [0085]-[0086]; para. [0091]), corresponding to original limitation “accessing a dataset including a plurality of data subsets, each of the data subsets corresponding to a feature of the dataset”. deLevie further discloses the portions/chunk (i.e., the claimed features of the dataset) as including separate or distinct portions/chunks/pages of titles and line numberings as well as columns, corresponding to “features of the dataset corresponding to various data types”, and consistent with paragraph [0020] (page 4-5) of Applicant’s original disclosure that describes different columns as representing different types of data, and as such, Examiner maintains that deLevie discloses the limitation.
Applicant’s arguments (Arguments, pg. 11, second para. – pg. 12) with respect to the claims and deLevie not disclosing limitation “training a machine learning (ML) model using the dataset including the one or more additional data subsets” (as opposed to limitation “training a machine learning (ML) model using the dataset” previously recited in claim 2) have been considered but are moot in light of new grounds of rejection with reference Liu as presented below.
Response to Amendment
The prior 35 U.S.C. 101 rejection of the claims is hereby withdrawn in light of amendments to the claims as well as arguments (2/19/26) presented by Applicant.
Claim Rejections - 35 USC § 103
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 (i.e., changing from AIA to pre-AIA ) 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.
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.
1. Claims 1-5, 7, 9, 10, 12-16 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over deLevie US 2025/0061279 A1 (“deLevie”) in view of Saxena US 2024/0330579 A1 (“Saxena”) and Liu et al US 2025/0209282 A1 (“Liu”)
Per claim 1, deLevie discloses a method comprising:
accessing a dataset including a plurality of data subsets, each of the data subsets corresponding to a feature of the dataset (para. [0058]-[0059]; para. [0068]; A request to divide text into one or more portions is received at 502…. The request may identify a body of text. The body of text may include one or more documents …, para. [0085]-[0086]; the enumerated source text passages may be divided into subsets …, para. [0136], documents/texts as including subsets/portions),
features of the dataset corresponding to various data types (multipage pages may be split into individual pages via a machine learning model. The machine learning model may be trained to group together portions of text on a multipage page. For instance, a caption page in a legal decision may include text in a column on the left that encompasses the parties, text in a column on the right that includes the case number, a title that follows lower on the page, and line numbering on the left. In such a configuration, the machine learning model may be trained to treat separately the text in the different columns …, para. [0061], text in different columns as example different data types);
analyzing data in one of the data subsets to determine a characteristic of the data (para. [0068]-[0069]; para. [0092]; para. [0113]-[0114]; para. [0123]; para. [0134]; the enumerated source text passages may be divided into subsets and analyzed in any suitable order, in sequence or in parallel. For example, the enumerated source text passages may be grouped by the chunk from which they were extracted. As another example, the enumerated source text passages may be re-grouped according to some other scheme, such as by topic determined …, para. [0136]);
selecting a prompt template from among a plurality of prompt templates for the one of the data subsets (the request may be analyzed to search for keywords or other indications that a particular text generation flow is desired…., para. [0069]; One or more prompt templates are determined at 408 based on the input text … different text generation flows may be associated with different prompt templates. Prompt templates may be selected from the prompt library based on the particular text generation flow, para. [0073]; para. [0074]);
generating a plurality of large language model prompts using the prompt template and the data from the one of the data subsets (para. [0021]; one or more prompts based on the prompt templates are determined.…, para. [0074]);
providing the plurality of large language model prompts to a large language model, the plurality of large language model prompts commanding the large language model to improve or generate additional data with respect to the data of the one of the data subsets (the text generation model 276 may be a large language model…., para. [0044]; The prompt template may be modified to determine a prompt by adding a portion of input text that characterizes the nature of the correspondence document to be generated…., para. [0074]; The one or more prompts are transmitted to a text generation modeling system …, para. [0075]; the one or more text response messages include one or more novel text portions generated by a text generation model ..., para. [0076]-[0077]; Parsing the text generation response message may involve, for instance, separating the novel text portion generated by the large language model from the rest of the completed text generation response message, para. [0138], novel text portions generated by text generation model as generated additional data); and
deLevie discloses creating one or more additional data subsets (para. [0076]-[0078]; para. [0138]), but does not explicitly disclose selecting a prompt template from among a plurality of prompt templates for the one of the data subsets based on the determined characteristic of the data of the one of the data subsets or creating one or more additional data subsets for the dataset based on responses of the large language model, each of the one or more additional data subsets corresponding to a new feature of the dataset
However, these features are taught by Saxena:
selecting a prompt template from among a plurality of prompt templates for the one of the data subsets based on the determined characteristic of the data of the one of the data subsets (a first prompt template can be associated with (and used to generate text for) a first type of webpage section (e.g., an announcement bar) while a second, different prompt template is associated with a second type of webpage section (e.g., a company description section). The available webpage section types offered by the website development system 120 can each have one or more associated prompt templates …, para. [0030]; The prompt template can be selected based on section type …, para. [0031])
creating one or more additional data subsets for the dataset based on responses of the large language model, each of the one or more additional data subsets corresponding to a new feature of the dataset (para. [0020]-[0021]; The LLM 130 uses the prompt to generate text 135. The website development system 120 can then add the generated text 135 to the website 105 …, para. [0022]; para. [0030], generated new text for a section/portion of the webpage as additional data subset)
deLevie in view of Saxena does not explicitly disclose training a machine learning (ML) model using the dataset including the one or more additional data subsets
However, this feature is taught by Liu (para. [0002]; the large language model 402 can receive a training dataset 408 comprising a large corpus of unlabeled data such as text, video, audio, and images. The large language model 402 can process the training dataset 408 in accordance with a natural language generation task 410…. In response to the natural language generation task 410, the large language model 402 can generate a first natural language output 412…., para. [0049]; The first natural language output 412 can be accordingly provided to the natural language generation system 404 along with the training dataset 408 to train a small language model 414 using an imitation learning protocol 416. …, para. [0050], training dataset and natural language output/text as dataset and additional data subset used in training ML model)
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Saxena with the method of deLevie in arriving at the missing features of deLevie, as well as to combine the teachings of Liu with the method of DeLevie in view of Saxena in arriving at the missing features of DeLevie in view of Saxena, because such combination would have resulted in quickly and efficiently providing text appropriate for the context in which the text will appear (Saxena, para. [0017]-[0018]), and in learning behaviors across various natural language tasks (Liu, para. [0050]).
Per claim 2, deLevie in view of Saxena and Liu discloses the method of claim 1,
Liu discloses performing one or more operations using the ML model (para. [0050]).
Per claim 3, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie discloses wherein the data from the one of the data subsets includes character strings and the plurality of large language model prompts include dividing each of the character strings into two or more sub-strings, wherein the one or more additional data subsets are created using the two or more sub-strings (identifying and enumerating the set of text passages may involve dividing the source documents into text chunks. The text chunks may be further divided into passages such as sentences, phrases, or paragraphs.…, para. [0035]-[0036]; para. [0076]-[0078]; para. [0138]).
Per claim 4, deLevie in view of Saxena and Liu discloses the method of claim 3,
deLevie discloses wherein a portion of the data of the one of the data subsets is provided to the large language model, the method further comprising: determining a string division rule based on the responses of the large language model (para. [0061]; para. [0136]; The text generation prompt is transmitted to a large language model for completion at 910. A text generation response message is received from the large language model at 912. The text generation response message is parsed at 914 to identify a novel text portion. Parsing the text generation response message may involve, for instance, separating the novel text portion generated by the large language model from the rest of the completed text generation response message, para. [0138]); and
dividing a remaining portion of the data of the one of the data subsets into two or more sub-strings using the string division rule (fig. 9, element 906, 914, 916; para. [0136]-[0138]; A determination is made at 916 as to whether to select one or more additional enumerated source text passages for analysis. …, para. [0139]).
Per claim 5, deLevie in view of Saxena and Liu discloses the method of claim 4,
deLevie discloses: wherein the method further comprises: determining whether a quality of the string division rule is sufficient for a machine learning model (para. [0084]); and
in response to determining the string division rule is not sufficient for the machine learning model, determining a new string division rule (para. [0084]).
Per claim 7, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie discloses wherein the plurality of large language model prompts are provided to the large language model in parallel (para. [0117]-[0118]; para. [0136]).
Per claim 9, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie discloses wherein data from the one of the data subsets is text data and the plurality of large language model prompts include requesting additional information regarding the text, wherein the one or more additional data subsets include the additional information from the large language model (para. [0072]; para. [0076]-[0078]; para. [0138]).
Per claim 10, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie discloses wherein data from the one of the data subsets is text data, wherein the one or more additional data subsets include information regarding subphrases of individual phrases in the data subsets (para. [0035]; para. [0072]; para. [0076]-[0078]; para. [0138]).
Per claim 12, deLevie discloses one or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause a system to perform operations, the operations comprising:
accessing a dataset including a plurality of data subsets, each of the data subsets corresponding to a feature of the dataset (para. [0058]-[0059]; para. [0068]; A request to divide text into one or more portions is received at 502…. The request may identify a body of text. The body of text may include one or more documents …, para. [0085]-[0086]; the enumerated source text passages may be divided into subsets …, para. [0136], documents/texts as including subsets/portions),
features of the dataset corresponding to various data types (multipage pages may be split into individual pages via a machine learning model. The machine learning model may be trained to group together portions of text on a multipage page. For instance, a caption page in a legal decision may include text in a column on the left that encompasses the parties, text in a column on the right that includes the case number, a title that follows lower on the page, and line numbering on the left. In such a configuration, the machine learning model may be trained to treat separately the text in the different columns …, para. [0061], text in different columns as example different data types);
analyzing data in one of the data subsets to determine a characteristic of the data (para. [0068]-[0069]; para. [0092]; para. [0113]-[0114]; para. [0123]; para. [0134]; the enumerated source text passages may be divided into subsets and analyzed in any suitable order, in sequence or in parallel. For example, the enumerated source text passages may be grouped by the chunk from which they were extracted. As another example, the enumerated source text passages may be re-grouped according to some other scheme, such as by topic determined …, para. [0136]);
selecting a prompt template from among a plurality of prompt templates for the one of the data subsets (the request may be analyzed to search for keywords or other indications that a particular text generation flow is desired…., para. [0069]; One or more prompt templates are determined at 408 based on the input text … different text generation flows may be associated with different prompt templates. Prompt templates may be selected from the prompt library based on the particular text generation flow, para. [0073]; para. [0074]);
generating a plurality of large language model prompts using the prompt template and the data from the one of the data subsets (para. [0021]; one or more prompts based on the prompt templates are determined.…, para. [0074]);
providing the plurality of large language model prompts to a large language model, the plurality of large language model prompts commanding the large language model to improve or generate additional data with respect to the data of the one of the data subsets (the text generation model 276 may be a large language model…., para. [0044]; The prompt template may be modified to determine a prompt by adding a portion of input text that characterizes the nature of the correspondence document to be generated…., para. [0074]; The one or more prompts are transmitted to a text generation modeling system …, para. [0075]; the one or more text response messages include one or more novel text portions generated by a text generation model ..., para. [0076]-[0077]; Parsing the text generation response message may involve, for instance, separating the novel text portion generated by the large language model from the rest of the completed text generation response message, para. [0138], novel text portions generated by text generation model as generated additional data); and
deLevie discloses creating one or more additional data subsets (para. [0076]-[0078]; para. [0138]), but does not explicitly disclose selecting a prompt template from among a plurality of prompt templates for the one of the data subsets based on the determined characteristic of the data of the one of the data subsets or creating one or more additional data subsets for the dataset based on responses of the large language model, each of the one or more additional data subsets corresponding to a new feature of the dataset
However, these features are taught by Saxena:
selecting a prompt template from among a plurality of prompt templates for the one of the data subsets based on the determined characteristic of the data of the one of the data subsets (a first prompt template can be associated with (and used to generate text for) a first type of webpage section (e.g., an announcement bar) while a second, different prompt template is associated with a second type of webpage section (e.g., a company description section). The available webpage section types offered by the website development system 120 can each have one or more associated prompt templates …, para. [0030]; The prompt template can be selected based on section type …, para. [0031])
creating one or more additional data subsets for the dataset based on responses of the large language model, each of the one or more additional data subsets corresponding to a new feature of the dataset (para. [0020]-[0021]; The LLM 130 uses the prompt to generate text 135. The website development system 120 can then add the generated text 135 to the website 105 …, para. [0022]; para. [0030], generated new text for a section/portion of the webpage as additional data subset)
deLevie in view of Saxena does not explicitly disclose training a machine learning (ML) model using the dataset including the one or more additional data subsets
However, this feature is taught by Liu (para. [0002]; the large language model 402 can receive a training dataset 408 comprising a large corpus of unlabeled data such as text, video, audio, and images. The large language model 402 can process the training dataset 408 in accordance with a natural language generation task 410…. In response to the natural language generation task 410, the large language model 402 can generate a first natural language output 412…., para. [0049]; The first natural language output 412 can be accordingly provided to the natural language generation system 404 along with the training dataset 408 to train a small language model 414 using an imitation learning protocol 416. …, para. [0050], training dataset and natural language output/text as dataset and additional data subset used in training ML model)
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Saxena with the media of deLevie in arriving at the missing features of deLevie, as well as to combine the teachings of Liu with the method of DeLevie in view of Saxena in arriving at the missing features of DeLevie in view of Saxena, because such combination would have resulted in quickly and efficiently providing text appropriate for the context in which the text will appear (Saxena, para. [0017]-[0018]), and in learning behaviors across various natural language tasks (Liu, para. [0050]).
Per claim 13, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 12,
Media claim 13 and method claim 2 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 13 is similarly rejected under the same rationale as applied above with respect to claim 2.
Per claim 14, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 12,
Media claim 14 and method claim 3 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 14 is similarly rejected under the same rationale as applied above with respect to claim 3.
Per claim 15, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 14,
Media claim 15 and method claim 4 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 15 is similarly rejected under the same rationale as applied above with respect to claim 4.
Per claim 16, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 15,
Media claim 16 and method claim 5 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 16 is similarly rejected under the same rationale as applied above with respect to claim 5.
Per claim 18, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 12,
Media claim 18 and method claim 7 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 18 is similarly rejected under the same rationale as applied above with respect to claim 7.
2. Claims 6 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over deLevie in view of Saxena and Liu as applied to claims 1 and 12 above, and further in view of Ghoche et al US 2024/0177172 A1 (“Ghoche”)
Per claim 6, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie in view of Saxena does not explicitly disclose wherein the plurality of large language model prompts includes clustering the data of the one of the data subsets, the method further comprising assigning an identifier for each cluster identified by the response of the large language model, wherein the one or more additional data subsets includes information regarding the clusters
However, this feature is taught by Ghoche (para. [0146]; para. [0170]-[0171])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Ghoche with the method of deLevie in view of Saxena and Liu in arriving at the missing features of deLevie in view of Saxena, because such combination would have resulted in provided responses to user queries within a desired threshold level of accuracy (Ghoche, para. [0064]).
Per claim 17, deLevie in view of Saxena and Liu discloses the one or more non-transitory computer-readable media of claim 12,
Media claim 17 and method claim 6 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 17 is similarly rejected under the same rationale as applied above with respect to claim 6.
3. Claims 8 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over deLevie in view of Saxena and Liu as applied to claims 1 and 12 above, and further in view of Rogynskyy et al US 2025/0045313 A1 (“Rogynskyy”)
Per claim 8, deLevie in view of Saxena and Liu discloses the method of claim 1,
deLevie in view of Saxena does not explicitly disclose replacing the one of the data subsets with one of the one or more additional data subsets
However, this feature is taught by Rogynskyy (para. [0476])
It would have been obvious to one of ordinary skill in the art before the effective filing of the invention to combine the teachings of Rogynskyy with the method of deLevie in view of Saxena and Liu in arriving at the missing features of deLevie in view of Saxena, because such combination would have resulted in conserving memory resources while maintaining a new state of the record object that can be used to generate responses to queries (Rogynskyy, para. [0476]).
Per claim 19, deLevie in view of Saxena discloses the one or more non-transitory computer-readable media of claim 12,
Media claim 19 and method claim 8 are related as media and the method of using same, with each claimed element's function corresponding to the claimed method step. Accordingly claim 19 is similarly rejected under the same rationale as applied above with respect to claim 8.
Allowable Subject Matter
Claim 11 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.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO 892 form.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to OLUJIMI A ADESANYA whose telephone number is (571)270-3307. The examiner can normally be reached Monday-Friday 8:30-5:00pm.
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.
/OLUJIMI A ADESANYA/Primary Examiner, Art Unit 2658
Prosecution Insights