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 .
Examiner called Applicant, Ran Wang, on May 07, 2026, to disclose overcoming the 35 USC 112, second paragraph issues introduced by the claim amendment, filed March 17, 2026. No agreement was reached to advance prosecution of the application.
REMARKS
On pages 9-10, Applicant argues amended independent claim 1 recites additional elements that contribute to an improvement to the machine learning technology and integrate the allegedly abstract idea into a practical application, making the claim patent- eligible under Step 2A, Prong Two. Applicant points to paragraphs [0011] and [0012] to support the asserted improvement and integration the abstract idea into a practical application. Applicant’s argument is persuasive and the 35 USC 101 rejection as applied to claims 1-20 is withdrawn.
On pages 10-11, Applicant’s argument by claim amendment to overcome the 35 USC 103 rejection in view of Gross et al. and Vangala et al. as applied to claims 1, 3, 5-8, 10, 12, 13, 14, 15, 17, 19, and 20 is persuasive. The 35 USC 103 rejection in view of Gross et al. and Vangala et al. as applied to claims 1, 3, 5-8, 10, 12, 13, 14, 15, 17, 19, and 20 is withdrawn.
Claims 1-20, filed March 17, 2026, are examined on the merits.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 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 applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1, line 12, the limitation of “reliability information” causes the claim as a whole to be vague and indefinite because the term “reliability” is not clearly defined in the claim and/or in the instant specification. The claim and/or specification does not provide the criteria to which to one of ordinary skill in the art to ascertain the reliability of information as claimed. One of ordinary skill in the art would not be able to determine the reliability of information based on the claim and/or the instant specification. Further, lines 13-14 recites “computing a weighted sum based on the number of edges and the reliability information”, wherein the application is not clear as to whether “the reliability information” is a value, number, or something else being used to calculated the weighted sum. The same issue is present in claims 8 and 15. Claims 2-7, 9-14, and 16-20 are rejected for being dependent from claim 1, 8, or 15, respectively.
Claim 1, line 18, the limitation of “high quality” causes the claim as a whole to be vague and indefinite because the term “quality” is not clearly defined in the claim and/or in the instant specification. The claim and/or specification does not provide the criteria to which to one of ordinary skill in the art to ascertain the quality of the data source as claimed. One of ordinary skill in the art would not be able to determine the quality of the data source based on the claim and/or the instant specification. Further, lines 16-19, suggests a comparison is being a performed, however, the application is not clear as whether the “quality” is a value, number, or something else being used for the comparison. The same issue is present in claims 8 and 15. Claims 2-7, 9-14, and 16-20 are rejected for being dependent from claim 1, 8, or 15, respectively.
Due to the vague and indefinite issues with the new limitations as discussed above, the new limitations have been attributed with the broadest reasonable interpretation (BRI) for the prior art rejection.
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.
Claim(s) 1, 3, 5-8, 10, 12, 13, and 14, 15, 17, 19, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Gross et al. (Gross hereafter, US 20250225402 A1) in view of Vangala et al. (Vangala hereafter, US 20240419918 A1) and Scheideler et al. (Scheideler hereafter, US 20190155926 A1).
Claim 1, Gross discloses a method comprising:
extracting a plurality of references ([0122], e.g. Documents from one or more sources may be accessed) from a plurality of data items received from a plurality of data sources ([0060], e.g. the LLM system 104 may utilize natural language processing (NLP) to extract keywords, sentiment analysis to determine the mood, and/or semantic parsing to extract the structure of the input text, and [0218], e.g. the processing device is configured to extract a first set of one or more “response claims” from the generated document. In this scenario, the response claims extracted by the processing devices consists of one or more sentences, each sentence comprising a statement, assertion, or declaration made by or about entities, events, concepts in a given context);
generating, by a processing device, a data structure ([0060], e.g. the LLM system 104 may utilize natural language processing (NLP) to extract keywords, sentiment analysis to determine the mood, and/or semantic parsing to extract the structure of the input text);
determining, based on the data structure, a plurality of scores ([0040], e.g. A contribution score or other indicator may be generated using such determination for respective items of content used for training) respectively associated with the plurality of data sources ([0261], e.g. estimating a contribution of a first item of content to the generated content based on a plurality of claims, wherein the plurality of claims comprise a first claim extracted from the generated content and a second claim extracted from the first item of content; generating feedback based at least in part on the estimated contribution of the first item of content to the generated content); and generating a training dataset for training a large language model (LLM) based on the plurality of data items and the plurality of scores ([0220], e.g. the processing device is configured to identify a second set of one or more “source claims” from the corpus of documents from which the LLM was directly trained or from documents from which the LLM was indirectly trained. The corpus generally includes thousands, if not millions, of documents with text, each document including one or more claims, namely factual claims and/or opinion claims. The second set of one or more source claims are claims extracted from the corpus based on their similarity to one or more response claims identified above).
However, Gross does not disclose the structure comprising a plurality of nodes and a plurality of edges, wherein the plurality of nodes are respectively associated with the plurality of data sources, and the plurality of edges are respectively associated with the plurality of references…wherein, for each node, determining the plurality of scores comprises: counting a number of edges in connection with the node; determining reliability information of the data source associated with the node; computing a weighted sum based on the number of edges and the reliability information; and normalizing the weight sum;… wherein data items from a high quality data source has greater influence than data items from a low quality data source on the training dataset.
Vangala discloses the structure comprising a plurality of nodes and a plurality of edges, wherein the plurality of nodes are respectively associated with the plurality of data sources, and the plurality of edges are respectively associated with the plurality of references ([0005], e.g. converting training graph data outputs from a data graph into a text data format that is readable by the LLM, the data graph having nodes and edges between the nodes, the nodes representing entities associated with an enterprise organization, and the edges representing relationships among the entities: generating a training set that comprises the converted training graph data outputs: training the LLM to receive graph data as an input prompt using the training set; and providing an extraction prompt to the LLM, the extraction prompt comprising syntax examples for the LLM to extract second graph data outputs from the data graph).
Vangala discloses data graphs often contain information that improves searches, predictions, recommendations, entity-entity lookups, clustering, and other processing scenarios. In some cases, an enterprise level data graph may have millions of nodes and billions of relationships ([0019]). One of ordinary skill in the art at the time prior to the effective filing date of the instant invention would have been motivated by Vangala to improve the method of Gross. Therefore, it would have been obvious for one of ordinary skill in the art to use method of Gross with the graph structure described by Vangala. The benefit would be to improve searches, predictions, recommendations, entity-entity lookups, clustering, and other processing scenarios.
However, Gross as modified does not disclose wherein, for each node, determining the plurality of scores comprises: counting a number of edges in connection with the node; determining reliability information of the data source associated with the node; computing a weighted sum based on the number of edges and the reliability information; and normalizing the weight sum;… wherein data items from a high quality data source has greater influence than data items from a low quality data source on the training dataset. Schedieler discloses wherein, for each node, determining the plurality of scores comprises: counting a number of edges in connection with the node; determining reliability information of the data source associated with the node; computing a weighted sum based on the number of edges and the reliability information; and normalizing the weight sum ([0071] In one implementation or embodiment similar to scoring model a), however, instead of counting edges, the sum of weights of the edges is taken as a criterion to distinguish between persistent nodes, branch nodes and leaf nodes. [0072] And another embodiment or implementation similar to the scoring model b), however, the sum of the weights along the path (instead of the length of the path) to the next persistent node is used to identify expired nodes);… wherein data items from a high quality data source has greater influence than data items from a low quality data source on the training dataset ([0083], e.g. Several or all scoring models are applied. A predetermined weight factor is applied to the result of each relevance degree value. The sum is normalized to the value of the range zero (0=not to be expired) to one (1=expired). If a normalized value of a certain node exceeds a predetermined value (e.g., 0.5, other values possible), the node is regarded to be expired and thus to be deleted).
Schedieler discloses an invention for continuous efficiency improvements ([0077]). One of ordinary skill in the art at the time before the effective filing date of the claimed invention would have been motivated by Schedieler to improve the system of Gross as modified. Therefore, it would have been obvious for one of ordinary skill in the art to use the system of Gross as modified with the teachings of Schedieler. The benefit would be to make continuous efficiency improvements.
Claim 3, Gross as modified discloses the LLM model corresponds to a generative artificial intelligence (AI) application for a semantic topic (Vangala, [0050], e.g. embeddings are generated for data graphs where the embeddings represent semantics of entities within an enterprise organization), the method further comprising: creating a link according to the semantic topic to indicate a relationship between two data sources (Vangala, [0075], e.g. the data graph is a heterogenous graph having nodes with different types, the entities include one or more of users, documents, emails, meetings, and conversations, and the relationships include one or more of document authorship, document modification, document sharing, meeting invites, linked data between documents).
Claim 5, Gross as modified disclose generating the training dataset comprises: selecting, from the plurality of data sources, one or more data sources associated with scores that satisfy a threshold (Gross, [0111], e.g. If the uncertainty is greater than a threshold); and sampling data items from the one or more data sources based on one or more sampling weights that correlate to the scores of the one or more data sources (Gross, [0111], e.g. A determination may be made as to how much/what percentage of the LLM output 502A can be attributed to the common reference source, which may in turn be used in determining the attribution assigned to the other sources (e.g., by weighting the contribution to the common reference source)).
Claim 6, Gross as modified discloses selecting, from a pool of data sources, the plurality of data sources that are relevant to the semantic topic (Vangala, [0085], e.g. step 708 further comprises selecting the one or more nodes as a subset of the plurality of nodes for the output data according to the user context for the target user and generating a summary of content of the one or more documents, the converted graph data output representing weights for the one or more documents according to the target user).
Claim 7, Gross as modified discloses receiving, from a client device, a query about the semantic topic; and deploying the generative AI application to generate a response to the query based on the LLM model (Vangala, [0040], e.g. the LLM 168 is able to generate queries for the node processors to retrieve graph data or documents based on an input prompt. Generally, the extraction prompt describes a structure of the data graph (e.g., nodes, edges, metadata, fields, etc.) and semantics for how a request to a node processor should be formatted).
Claims 8, 10, 12, 13, and 14, 15, 17, 19, and 20, directed to a system and non-transitory computer-readable medium for implementing the above cited method. Gross as modified discloses a system and non-transitory computer-readable medium (Figure 1) for implementing the above cited method. Therefore, claims 8, 10, 13, and 14, 15, 17, and 20 are rejected for the same citations and rationale.
Claim(s) 2, 9, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Gross et al. (Gross hereafter, US 20250225402 A1) in view of Vangala et al. (Vangala hereafter, US 20240419918 A1) and Scheideler et al. (Scheideler hereafter, US 20190155926 A1), as applied to claims 1, 3, 5-8, 10, 12, 13, and 14, 15, 17, 19, and 20 above, in further view of Kinoshita (US 20040073925 A1).
Claim 2, Gross discloses the claimed invention except for detecting a data media format of the plurality of data items; selecting a media conversion application based on the data media format, wherein the media conversion application is to convert a data item from the data media format to another format; and deploying the media conversion application to the plurality of data items. Kinoshita discloses detecting a data media format of the plurality of data items; selecting a media conversion application based on the data media format, wherein the media conversion application is to convert a data item from the data media format to another format; and deploying the media conversion application to the plurality of data items ([0013], e.g. when a content delivery request identifying a user and a content is received from a content owner, searching the first memory for the user to find corresponding user information; selecting a most suitable one from the format conversion programs based on the content replaying environment included in the found user information, wherein the most suitable format conversion program provides a format most suitable for the content replaying environment of the user; converting the content into the format according to the most suitable format conversion program to produce a format-converted content for the user; and delivering the format-converted content to the user through the network).
Kinoshita discloses the optimum content format is automatically provided to the content user without burdening the content owner and the content user ([0014]). One of ordinary skill in the art at the time prior to the effective filing date of the instant invention would have been motivated by Kinoshita to improve the method of Gross as modified. Therefore, it would have been obvious for one of ordinary skill in the art to use the method of Gross as modified with the format application selector of Kinoshita. The benefit would be to automatically provide content to the user without burdening the content owner and the content user.
Claims 9 and 16, directed to a system and non-transitory computer-readable medium for implementing the above cited method. Gross as modified discloses a system and non-transitory computer-readable medium (Figure 1) for implementing the above cited method. Therefore, claims 9 and 16 are rejected for the same citations and rationale.
PERTINENT PRIOR ART
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Yu et al. (CN 115526339 A) parameter, determining the quality of each node, then selecting the quality of the edge satisfy of the preset condition as the edge node of the first wheel federated learning For example, the performance parameter comprises sample data amount as an example, the sample data is directly used as the quality of the node, or the sample data to a certain mathematical calculation after the value as the mass, the data calculation can be normalization processing and so on, the specific processing mode of the data calculation is not limited. for example, performance parameter comprises sample data and GPU performance parameter, at this time, the quality of the node can be the sample number and GPU performance parameter indication of the GPU performance data according to a certain weight sum, so as to obtain the quality of the edge node (Page 11)
CONCLUSION
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.
Patent applicants with problems or questions regarding electronic images that can be viewed in the Patent Application Information Retrieval system (PAIR) can now contact the USPTO's Patent Electronic Business Center (Patent EBC) for assistance. Representatives are available to answer your questions daily from 6 am to midnight (EST). The toll free number is (866) 217-9197. When calling please have your application serial or patent number, the type of document you are having an image problem with, the number of pages and the specific nature of the problem. The Patent Electronic Business Center will notify applicants of the resolution of the problem within 5-7 business days. Applicants can also check PAIR to confirm that the problem has been corrected. The USPTO's Patent Electronic Business Center is a complete service center supporting all patent business on the Internet. The USPTO's PAIR system provides Internet-based access to patent application status and history information. It also enables applicants to view the scanned images of their own application file folder(s) as well as general patent information available to the public.
For all other customer support, please call the USPTO Call Center (UCC) at 800-786-9199. The USPTO's official fax number is 571-272-8300.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to C. Dune Ly, whose telephone number is (571) 272-0716. The examiner can normally be reached on Monday-Friday from 8 A.M. to 4 PM ET.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Tony Mahmoudi, can be reached on 571-272-4078.
/Cheyne D Ly/
Primary Examiner, Art Unit 2152
6/5/2026
Prosecution Insights