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 .
In response to Applicant’s claims filed on March 11, 2026, claims 1-21 are now pending for examination in the application.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-patentable subject matter. The claims are directed to an abstract idea without significantly more.
Claim 1-21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The judicial exception is not integrated into a practical application. The claims do not include additional elements that are sufficient to amount to significantly more than judicial exception. The eligibility analysis in support of these findings is provided below, on Claim Rejections - 35 USC 101 accordance with the "2019 Revised Patent Subject Matter Eligibility Guidance" (published on 1/7/2019 in Fed, Register, Vol. 84, No. 4 at pgs. 50-57, hereinafter referred to as the "2019 PEG").
Step 1. in accordance with Step 1 of the eligibility inquiry (as explained in MPEP 2106), it is first noted the claim methods (claims 1-11), storage medium (claim 12-16), and system (claims 17-20) are directed to one of the eligible categories of subject matter and therefore satisfies Step 1.
Step 2A. In accordance with Step 2A, prong one of the 2019 PEG, it is noted that the independent claims recite an abstract idea falling within the Mental Processes & Mathematical Concepts enumerated groupings of abstract ideas set forth in the 2019 PEG. Examiner is of the position that independent claims 1, 8, and 15 are directed towards the Mathematical Concepts Grouping of Abstract Ideas.
Independent claim(s) 1 and 8 recites the following limitations directed towards a Mental Processes & Mathematical Concepts:
generating a first prompt instructing a generative language model (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a prompt)
to analyze the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to analyze a report)
and generate a plurality of custom queries configured
generating a second prompt instructing the generative language model(The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a prompt)
to generate a draft custom report that emulates the one or more example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a custom report);
evaluating the draft custom report against the one or more example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to evaluate a report)
to identify missing information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify missing information) and
generate additional queries to obtain the missing information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a query);
iteratively repeating the query processing and report generation steps using the additional queries until the custom report substantially replicates the structure and content organization of the one or more example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to replicate a structure).
Step 2A. In accordance with Step 2A, prong two of the 2019 PEG, the judicial exception is not integrated into a practical application because of the recitation in claim(s) 1 and 8:
One or more processors (i.e., as a generic processor/component performing a generic computer function);
one or more memory storage devices storing instructions thereon, which, when executed by the one or more processors (i.e., as a generic processor/component performing a generic computer function) including:
receiving, as input to a computing system, a set of example financial reports and a selected task (recites insignificant extra solution activity that amounts to mere data gathering);
processing the plurality of custom queries through a query processing pipeline to obtain query results from one or more data sources (recites insignificant extra solution activity that amounts to obtaining results data);
outputting the custom report in a format that maintains the structure and content organization of the example financial reports (recites insignificant extra solution activity that amounts to outputting a report); and
Step 2B. Similar to the analysis under 2A Prong Two, the claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Because the additional elements of the independent claims amount to insignificant extra solution activity and/or mere instructions, the additional elements do not add significantly more to the judicial exception such that the independent claims as a whole would be patent eligible.
Independent claim(s) 15 recites the following limitations directed towards a Mental Processes:
generating a first prompt for a generative language model, the first prompt including an instruction directing the generative language model to analyze the set of example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a prompt)
to identify content structure and formatting characteristics (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify a structure), and
to generate a plurality of custom queries configured to retrieve information of a type contained in the set of example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a query);
filtering the retrieved documents to remove irrelevant information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to filtering documents), and
reranking the filtered documents based on relevance to the custom query (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to reranking documents);
generating a second prompt for the generative language model, the second prompt including an instruction (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a prompt)
to generate a draft custom report that emulates the format and content structure of the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a report);
generating a third prompt for the generative language model (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a prompt),
the third prompt including an instruction to evaluate the draft custom report against the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to evaluating a report)
to identify missing information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify missing information) and
generate one or more additional queries (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generating a query)
iteratively repeating the query processing and report generation steps using the one or more additional queries until the custom report substantially replicates the format and content structure of the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to replicate a structure).
Step 2A. In accordance with Step 2A, prong two of the 2019 PEG, the judicial exception is not integrated into a practical application because of the recitation in claim(s) 15:
receiving, as input to a computing system, a set of example financial reports and a selected task associated with a predefined workflow (recites insignificant extra solution activity that amounts to mere data gathering);
processing the plurality of custom queries generated by the generative language model through a query processing pipeline to obtain a plurality of query results (recites insignificant extra solution activity that amounts to obtain query results),
retrieving relevant documents from one or more data sources (recites insignificant extra solution activity that amounts to retrieving data),
to synthesize the plurality of query results (recites insignificant extra solution activity that amounts to synthesizing data)
outputting the custom report in a structured format that maintains the specific formatting and content organization of the example financial reports (recites insignificant extra solution activity that amounts to outputting a report); and
Step 2B. Similar to the analysis under 2A Prong Two, the claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Because the additional elements of the independent claims amount to insignificant extra solution activity and/or mere instructions, the additional elements do not add significantly more to the judicial exception such that the independent claims as a whole would be patent eligible.
Therefore, independent claim(s) 1, 8, 15 is/are rejected under 35 U.S.C. 101.
With respect to claim(s) 2 and 9:
Step 2A, prong one of the 2019 PEG:
Examiner is of the position the dependent claim is directed toward additional elements.
Step 2A Prong Two Analysis:
wherein the selected task corresponds to a predefined workflow comprising a plurality of task-based queries, and wherein processing the plurality of custom queries further comprises processing the task-based queries to obtain additional query results that are combined with the custom query results when generating the draft custom report (recites insignificant extra solution activity that amounts to obtaining data).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 3 and 10:
Step 2A, prong one of the 2019 PEG:
augmenting each custom query with domain-specific knowledge (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to augmenting a query);
filtering the retrieved documents to remove irrelevant information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to filtering documents); and
ranking the filtered documents based on relevance to the custom query (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to ranking documents).
Step 2A Prong Two Analysis:
retrieving relevant documents from the one or more data sources (recites insignificant extra solution activity that amounts to retrieving data).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 4 and 11:
Step 2A, prong one of the 2019 PEG:
generating a third prompt instructing the generative language model (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a prompt)
to compare the draft custom report with the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to compare a report);
identifying content gaps where information present in the example financial reports is missing from the draft custom report (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify a content gap); and
automatically generating the additional queries based on the identified content gaps (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate queries).
Step 2A Prong Two Analysis:
This judicial exception is not integrated into a practical application because there are no
additional elements to provide practical application.
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 5 and 12:
Step 2A, prong one of the 2019 PEG:
wherein the example financial reports comprise reports having different structural formats, and wherein the first prompt further instructs the generative language model to:
identify common structural elements across the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify a common element); and
determine content types and presentation formats used in the example financial reports to guide generation of the custom queries (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to determine a content type).
Step 2A Prong Two Analysis:
This judicial exception is not integrated into a practical application because there are no
additional elements to provide practical application.
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 6 and 13:
Step 2A, prong one of the 2019 PEG:
wherein the iterative repeating step comprises:
determining a convergence metric by comparing the custom report with the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to determine a metric); and
continuing the iterative process when the convergence metric indicates insufficient similarity (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to continuing a process); and
terminating the iterative process when the convergence metric exceeds a predetermined threshold indicating substantial replication of the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to terminating a process).
Step 2A Prong Two Analysis:
This judicial exception is not integrated into a practical application because there are no
additional elements to provide practical application.
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 7 and 14:
Step 2A, prong one of the 2019 PEG:
customizing the query processing pipeline based on the selected role-based workflow to apply role-specific filtering rules and data source prioritization (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to customizing a pipeline).
Step 2A Prong Two Analysis:
receiving user selection of a role-based workflow from a plurality of available workflows, wherein each workflow is associated with different types of financial analysis tasks (recites insignificant extra solution activity that amounts to mere data gathering).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 16:
Step 2A, prong one of the 2019 PEG:
wherein the predefined workflow comprises a plurality of task-based queries associated with the selected task, and wherein processing the plurality of custom queries further comprises:
combining the task-based query results with the query results from the custom queries when generating the second prompt for synthesizing the draft custom report (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to combining results).
Step 2A Prong Two Analysis:
processing the plurality of task-based queries through the query processing pipeline to obtain task-based query results (recites insignificant extra solution activity that amounts to obtaining data).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 17:
Step 2A, prong one of the 2019 PEG:
wherein the domain-specific knowledge graph information comprises financial entity relationships, industry classifications, and regulatory frameworks, and wherein augmenting each custom query comprises:
identifying financial entities and concepts within the custom query (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identifying an entity);
expanding the custom query to include the related entities and contextual information (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to expanding a query).
Step 2A Prong Two Analysis:
retrieving related entities and contextual information from the domain-specific knowledge graph (recites insignificant extra solution activity that amounts to retrieving data).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 18:
Step 2A, prong one of the 2019 PEG:
wherein filtering the retrieved documents to remove irrelevant information comprises:
applying machine learning-based noise reduction algorithms to identify and remove documents that do not match the content requirements of the selected task (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify and remove documents);
scoring each document based on relevance to the custom query and the selected task (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to scoring a document); and
retaining only documents that exceed a predetermined relevance threshold (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to retain a document).
Step 2A Prong Two Analysis:
This judicial exception is not integrated into a practical application because there are no
additional elements to provide practical application.
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 19:
Step 2A, prong one of the 2019 PEG:
wherein the example financial reports comprise a plurality of reports having different formatting styles and content organizations, and wherein the first prompt further instructs the generative language model to:
identify common structural elements across the plurality of example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to identify structural elements);
determine section headings, content types, and presentation formats used in the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to determine formats); and
generate the plurality of custom queries to retrieve information suitable for populating each identified structural element (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a query).
Step 2A Prong Two Analysis:
This judicial exception is not integrated into a practical application because there are no
additional elements to provide practical application.
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 20:
Step 2A, prong one of the 2019 PEG:
wherein the iterative repeating step comprises:
processing the one or more additional queries through the query processing pipeline to obtain additional query results;
generating a fourth prompt for the generative language model (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a prompt)
to generate an updated custom report (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to generate a report);
comparing the updated custom report with the example financial reports (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to comparing a report)
to determine a convergence metric (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to determine a metric); and
terminating the iterative process when the convergence metric indicates that the updated custom report substantially matches the format and content structure of the example financial reports within a predetermined threshold (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to terminating a process).
Step 2A Prong Two Analysis:
to synthesize the additional query results with previously obtained query results (recites insignificant extra solution activity that amounts to synthesizing data)
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
With respect to claim(s) 21:
Step 2A, prong one of the 2019 PEG:
wherein the query processing pipeline is configured to:
route different custom queries to different data sources based on the type of information required (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to routing a query); and
aggregate results from multiple data sources for individual custom queries when comprehensive information is needed (The limitation recites a mental process of observation and/or evaluation capable of being performed by the human mind by using computer as a tool to aggregating results).
Step 2A Prong Two Analysis:
wherein the one or more data sources comprise at least two of: stock exchange data feeds, financial news outlets, legal decision databases, market research repositories, credit rating agency databases, regulatory filing systems, and company financial statements (recites insignificant extra solution activity that amounts to retriving data).
Step 2B Analysis:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claim is not patent eligible.
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) 1-2, 5, 8-9, and 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) in view of Paulett et al. (US Pub. No. 20240347156).
With respect to claim 1, Campos et al. discloses a computer-implemented method for generating a custom financial report based on one or more example financial reports, the method comprising:
receiving, as input to a computing system, a set of example financial reports and a selected task (Column 4 Lines 5-32 discloses obtain and/or access documents forming a corpus of electronic documents, and/or of a set of exemplary documents and Column 4 Lines 5-32 discloses original documents may include financial reports, financial records, and/or other financial documents);
generating a first prompt instructing a generative language model to analyze the example financial reports and generate a plurality of custom queries configured to retrieve information of a type contained in the one or more example financial reports (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133);
processing the plurality of custom queries through a query processing pipeline to obtain query results from one or more data sources (Column 10 Lines 50-67 and Column 11 Lines 1-4 discloses a convolutional neural network may have been trained and used to classify (pixelated) image data as characters, photographs, diagrams, media content, and/or other types of information. In some implementations, the extracted information may have been extracted by a document analysis process that uses a pipeline of steps for object detection, object recognition, and/or object classification). Campos et al. does not disclose generating a second prompt …; evaluating the draft custom report…; iteratively repeating the query processing…; and outputting the custom report.
However, Paulett et al. discloses generating a second prompt instructing the generative language model to synthesize the query results to generate a draft custom report that emulates the one or more example financial reports (Paragraph 78 discloses determining the template and/or macro can include using LLM-based methods. In a specific example, the input determination model can use a prompt-based approach. In an illustrative example, the input determination model can combine a large language model (LLM) with an information retrieval system (e.g., a semantic search engine), wherein the task of the LLM (and prompt) can be to generate the best possible template search string);
evaluating the draft custom report against the one or more example financial reports to identify missing information and generate additional queries to obtain the missing information (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model, and/or any other model to look for errors (e.g., report generation errors) and Paragraph 114 discloses missing information);
iteratively repeating the query processing and report generation steps using the additional queries until the custom report substantially replicates the structure and content organization of the one or more example financial reports (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model and Paragraph 91 discloses generating a report can include replicating a field into another section of a report (e.g., wherein each field points to the same value)); and
outputting the custom report in a format that maintains the structure and content organization of the example financial reports (Paragraph 95 discloses generating outputs (e.g., text, data, etc.) based on the set of inputs (e.g., unstructured inputs), which can include expanding the information contained within the inputs (e.g., determining a full form version of a shorthand for a finding entered by a radiologist, adding context for clarification, etc.), modifying the format (e.g., language, language style, data format, etc.) of the inputs, and/or otherwise generating outputs).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model with Paulett et al.’s radiology reporting. This would have facilitated using language modeling in generating a report using post processing to adjust a draft report to finalization.
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 2, Campos et al. teaches the method of claim 1, wherein the selected task corresponds to a predefined workflow comprising a plurality of task-based queries, and wherein processing the plurality of custom queries further comprises processing the task-based queries to obtain additional query results that are combined with the custom query results when generating the draft custom report (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 5, Campos et al. teaches the method of claim 1, wherein the example financial reports comprise reports having different structural formats, and wherein the first prompt further instructs the generative language model to:
identify common structural elements across the example financial reports (Column 9 Lines 29-46 discloses Large language model 133 may be able to determine and/or use whether information is formatted in a column, or a row, or a table. Accordingly, information elements in a column, or a row, or a table may be contextually and/or semantically linked and/or otherwise connected such that large language model 133 may extract information from a particular document based on knowledge of the formatted information in the particular document. In some implementations, model component 112 may be configured to obtain and/or present replies provided by large language model 133 to queries and/or prompts); and
determine content types and presentation formats used in the example financial reports to guide generation of the custom queries (Column 9 Lines 29-46 discloses Large language model 133 may be able to determine and/or use whether information is formatted in a column, or a row, or a table. Accordingly, information elements in a column, or a row, or a table may be contextually and/or semantically linked and/or otherwise connected such that large language model 133 may extract information from a particular document based on knowledge of the formatted information in the particular document. In some implementations, model component 112 may be configured to obtain and/or present replies provided by large language model 133 to queries and/or prompts).
With respect to claim 8, Campos et al. discloses a system for generating a custom financial report that has a format based on one or more example financial reports, the system comprising:
one or more processors (See Fig. 1); and
one or more memory storage devices (See Fig. 1) storing instructions thereon, which, when executed by the one or more processors, cause the system to perform operations comprising:
receiving, as input to a computing system, a set of example financial reports and a selected task (Column 4 Lines 5-32 discloses obtain and/or access documents forming a corpus of electronic documents, and/or of a set of exemplary documents and Column 4 Lines 5-32 discloses original documents may include financial reports, financial records, and/or other financial documents);
generating a first prompt instructing a generative language model to analyze the example financial reports and generate a plurality of custom queries configured to retrieve information of a type contained in the example financial reports (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133);
processing the plurality of custom queries through a query processing pipeline to obtain query results from one or more data sources (Column 10 Lines 50-67 and Column 11 Lines 1-4 discloses a convolutional neural network may have been trained and used to classify (pixelated) image data as characters, photographs, diagrams, media content, and/or other types of information. In some implementations, the extracted information may have been extracted by a document analysis process that uses a pipeline of steps for object detection, object recognition, and/or object classification). Campos et al. does not disclose generating a second prompt …; evaluating the draft custom report…; iteratively repeating the query processing…; and outputting the custom report.
However, Paulett et al. discloses generating a second prompt instructing the generative language model to synthesize the query results to generate a draft custom report that emulates the one or more example financial reports (Paragraph 78 discloses determining the template and/or macro can include using LLM-based methods. In a specific example, the input determination model can use a prompt-based approach. In an illustrative example, the input determination model can combine a large language model (LLM) with an information retrieval system (e.g., a semantic search engine), wherein the task of the LLM (and prompt) can be to generate the best possible template search string);
evaluating the draft custom report against the one or more example financial reports to identify missing information and generate additional queries to obtain the missing information (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model, and/or any other model to look for errors (e.g., report generation errors) and Paragraph 114 discloses missing information);
iteratively repeating the query processing and report generation steps using the additional queries until the custom report substantially replicates the structure and content organization of the one or more example financial reports (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model and Paragraph 91 discloses generating a report can include replicating a field into another section of a report (e.g., wherein each field points to the same value)); and
outputting the custom report in a format that maintains the structure and content organization of the example financial reports (Paragraph 95 discloses generating outputs (e.g., text, data, etc.) based on the set of inputs (e.g., unstructured inputs), which can include expanding the information contained within the inputs (e.g., determining a full form version of a shorthand for a finding entered by a radiologist, adding context for clarification, etc.), modifying the format (e.g., language, language style, data format, etc.) of the inputs, and/or otherwise generating outputs).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model with Paulett et al.’s radiology reporting. This would have facilitated using language modeling in generating a report using post processing to adjust a draft report to finalization.
With respect to claim 9, it is rejected on grounds corresponding to above rejected claim 2, because claim 9 is substantially equivalent to claim 2.
With respect to claim 12, it is rejected on grounds corresponding to above rejected claim 5, because claim 12 is substantially equivalent to claim 5.
Claim(s) 3, 10, 15-16, 19, and 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) in further view of Tomkins et al. (US Pub. No. 20210295822).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 3, Campos et al. teaches the method of claim 1, wherein the query processing pipeline comprises:
retrieving relevant documents from the one or more data sources (Column 10 Lines 13-32 discloses determine one or more documents from a corpus of electronic documents, the one or more documents being relevant to a particular query and/or a particular corresponding reply);
filtering the retrieved documents to remove irrelevant information (Column 10 Lines 13-32 discloses if a particular reply is based on information from one or more sections of a document, relevance component 110 may notify a user thereof. In some implementations, relevance component 110 may be configured to provide provenance for the contents of replies to queries). Campos et al. as modified by Paulett et al. does not disclose augmenting each custom query …; ranking the filtered documents.
However, Tomkins et al. discloses augmenting each custom query with domain-specific knowledge (Paragraph 289 discloses some embodiments may access a second ontology 1722 labeled with the domain “legal” to replace or augment one or more n-grams of the updated query with n-grams from the second ontology 1722);
ranking the filtered documents based on relevance to the custom query (Paragraph 142 discloses some embodiments may update a ranking of documents retrieved using the term “aFib” such that documents retrieved using an expanded query having the n-gram “Non-Valvular Atrial Fibulation” shown in the box 832 is assigned a greater priority in a search ranking).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with Tomkins et al.’s natural language model generation. This would have facilitated retrieving meaningful information for a query using natural language understanding.
With respect to claim 10, it is rejected on grounds corresponding to above rejected claim 3, because claim 10 is substantially equivalent to claim 3.
With respect to claim 15, Campos et al. discloses a computer-implemented method for generating a custom financial report that replicates a format and content structure of a set of example financial reports, the method comprising:
receiving, as input to a computing system, a set of example financial reports and a selected task associated with a predefined workflow (Column 4 Lines 5-32 discloses obtain and/or access documents forming a corpus of electronic documents, and/or of a set of exemplary documents and Column 4 Lines 5-32 discloses original documents may include financial reports, financial records, and/or other financial documents);
generating a first prompt for a generative language model, the first prompt including an instruction directing the generative language model to analyze the set of example financial reports to identify content structure and formatting characteristics, and to generate a plurality of custom queries configured to retrieve information of a type contained in the set of example financial reports (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133);
processing the plurality of custom queries generated by the generative language model through a query processing pipeline to obtain a plurality of query results, wherein the query processing pipeline includes reply). Campos et al. does not disclose generating a second prompt …; generating a third prompt …; iteratively repeating the query processing…; and outputting the custom report.
However, Paulett et al. discloses generating a second prompt for the generative language model, the second prompt including an instruction to synthesize the plurality of query results to generate a draft custom report that emulates the format and content structure of the example financial reports (Paragraph 78 discloses determining the template and/or macro can include using LLM-based methods. In a specific example, the input determination model can use a prompt-based approach. In an illustrative example, the input determination model can combine a large language model (LLM) with an information retrieval system (e.g., a semantic search engine), wherein the task of the LLM (and prompt) can be to generate the best possible template search string);
generating a third prompt for the generative language model, the third prompt including an instruction to evaluate the draft custom report against the example financial reports to identify missing information and generate one or more additional queries to obtain the missing information (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model, and/or any other model to look for errors (e.g., report generation errors) and Paragraph 114 discloses missing information);
iteratively repeating the query processing and report generation steps using the one or more additional queries until the custom report substantially replicates the format and content structure of the example financial reports (Paragraph 114 discloses the generated radiology report can be rerun (e.g., optionally alongside the original input data) through the same language model, a different language model and Paragraph 91 discloses generating a report can include replicating a field into another section of a report (e.g., wherein each field points to the same value)); and
outputting the custom report in a structured format that maintains the specific formatting and content organization of the example financial reports (Paragraph 95 discloses generating outputs (e.g., text, data, etc.) based on the set of inputs (e.g., unstructured inputs), which can include expanding the information contained within the inputs (e.g., determining a full form version of a shorthand for a finding entered by a radiologist, adding context for clarification, etc.), modifying the format (e.g., language, language style, data format, etc.) of the inputs, and/or otherwise generating outputs).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al. with Paulett et al. This would have facilitated using language modeling in generating a report using post processing to adjust a draft report to finalization.
Campos et al. as modified by Paulett et al. does not disclose
However, Tomkins et al. discloses augmenting each custom query with domain-specific knowledge (Paragraph 289 discloses some embodiments may access a second ontology 1722 labeled with the domain “legal” to replace or augment one or more n-grams of the updated query with n-grams from the second ontology 1722).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with Tomkins et al.’s natural language model generation. This would have facilitated retrieving meaningful information for a query using natural language understanding.
The Campos et al. reference as modified by Paulett et al. and Tomkins et al. teaches all the limitations of claim 15. With respect to claim 16, Campos et al. teaches the method of claim 15, wherein the predefined workflow comprises a plurality of task-based queries associated with the selected task, and wherein processing the plurality of custom queries further comprises:
processing the plurality of task-based queries through the query processing pipeline to obtain task-based query results (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133).; and
combining the task-based query results with the query results from the custom queries when generating the second prompt for synthesizing the draft custom report (Column 9 Lines 47-66 discloses provide queries as prompts to large language model 133. In some implementations, interface component 114 may be configured to obtain replies to queries from large language model 133).
The Campos et al. reference as modified by Paulett et al. and Tomkins et al. teaches all the limitations of claim 15. With respect to claim 19, Paulett et al. teaches the method of claim 15, wherein the example financial reports comprise a plurality of reports having different formatting styles and content organizations, and wherein the first prompt further instructs the generative language model to:
identify common structural elements across the plurality of example financial reports (Column 9 Lines 26-46 discloses the training documents may include financial documents, including but not limited to bank statements, insurance documents, mortgage documents, loan documents, and/or other financial documents. Large language model 133 may be able to determine and/or use whether information is formatted in a column, or a row, or a table. Accordingly, information elements in a column, or a row, or a table may be contextually and/or semantically linked and/or otherwise connected such that large language model 133 may extract information from a particular document based on knowledge of the formatted information in the particular document);
determine section headings, content types, and presentation formats used in the example financial reports (Column 9 Lines 26-46 discloses the training documents may include financial documents, including but not limited to bank statements, insurance documents, mortgage documents, loan documents, and/or other financial documents. Large language model 133 may be able to determine and/or use whether information is formatted in a column, or a row, or a table. Accordingly, information elements in a column, or a row, or a table may be contextually and/or semantically linked and/or otherwise connected such that large language model 133 may extract information from a particular document based on knowledge of the formatted information in the particular document); and
generate the plurality of custom queries to retrieve information suitable for populating each identified structural element (Column 9 Lines 26-46 discloses the training documents may include financial documents, including but not limited to bank statements, insurance documents, mortgage documents, loan documents, and/or other financial documents. Large language model 133 may be able to determine and/or use whether information is formatted in a column, or a row, or a table. Accordingly, information elements in a column, or a row, or a table may be contextually and/or semantically linked and/or otherwise connected such that large language model 133 may extract information from a particular document based on knowledge of the formatted information in the particular document).
The Campos et al. reference as modified by Paulett et al. and Tomkins et al. teaches all the limitations of claim 15. With respect to claim 21, Campos et al. teaches the method of claim 15, wherein the one or more data sources comprise at least two of: stock exchange data feeds, financial news outlets, legal decision databases, market research repositories, credit rating agency databases, regulatory filing systems, and company financial statements (Column 9 Lines 26-46 discloses the training documents may include financial documents, including but not limited to bank statements, insurance documents, mortgage documents, loan documents, and/or other financial documents;
wherein the query processing pipeline is configured to:
route different custom queries to different data sources based on the type of information required (interface component 114 may be configured to obtain replies to queries from large language model 133. For example, a user may enter a query to cause large language model 133 to extract the net amount spent in a particular week from exemplary electronic source document 30 in FIG. 3A or particular document 43a in FIG. 4 (e.g., based on the difference between beginning and ending balance, here “$160”)); and
aggregate results from multiple data sources for individual custom queries when comprehensive information is needed (Column 11 Lines 55-66 discloses external resources 120 may include sources of information outside of system 100, external entities participating with system 100, and/or other resources. In some implementations, external resources 120 may include a provider of documents, including but not limited to electronic source documents 123, from which system 100 and/or its components (e.g., source component 108) may obtain documents).
Claim(s) 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) and Tomkins et al. (US Pub. No. 20210295822) in further view of Mondlock et al. (US Patent No. 12079570).
The Campos et al. reference as modified by Paulett et al. and Tomkins et al. teaches all the limitations of claim 15. With respect to claim 17, Campos et al. as modified by Paulett et al. and Tomkins et al. does not disclose identifying financial entities and concepts within the custom query.
However, Mondlock et al. teaches the method of claim 15, wherein the domain-specific knowledge graph information comprises financial entity relationships, industry classifications, and regulatory frameworks, and wherein augmenting each custom query comprises:
identifying financial entities and concepts within the custom query (Column 7 Lines 37-52 discloses the user query may ask for a financial summary of Acme Corp., and the intent classification module 126 may classify the intent of the user query as public company financial information);
retrieving related entities and contextual information from the domain-specific knowledge graph (Column 9 Lines 37-42 discloses the external data sources 160A-160N may include knowledge graphs 216A-216N. The knowledge graphs 216A-216N may include a GraphDB, Virtuoso, or any other suitable knowledge graph. The knowledge graphs 216A-216N may host text and/or data); and
expanding the custom query to include the related entities and contextual information (Column 19 Lines 61-67 discloses the external data sources 160A-160N may include knowledge graphs 216A-216N. The knowledge graphs 216A-216N may include a GraphDB, Virtuoso, or any other suitable knowledge graph. The knowledge graphs 216A-216N may host text and/or data).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting and Tomkins et al.’s natural language model generation with Mondlock et al.’s generative AI pipeline. This would have facilitated a customizable answering of a query using a large language model.
Claim(s) 4 and 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) in further view of De Ridder (US Pub. No. 20210271823).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 4, Paulett et al. teaches the method of claim 1, wherein evaluating the draft custom report comprises:
identifying content gaps where information present in the example financial reports is missing from the draft custom report (Paragraph 114 discloses Examples of input errors can include: discrepancies between two or more sources of information, errors in the output of a findings model (e.g., errors in a speech to text operation), errors input by the radiologist, missing information (e.g., a finding from a previous exam for the patient a radiologist forgot to mention while completing a current radiology report for the patient), incorrect information, grammatical information, and/or any other errors); and
automatically generating the additional queries based on the identified content gaps (Paragraph 114 discloses Examples of report generation errors include: contradictions within the generated outputs themselves; contradictions, hallucinations, and/or significant missed findings relative to the original input data; duplicated and/or partially duplicated concepts and/or sentences; incorrect language (e.g., gibberish boilerplate language); errors related to numbers, measurements, and/or dates; errors related to similar anatomical structures (different spine levels, different metacarpals or metatarsals, etc.); mixing and/or combining similar findings; findings with somewhat different severity levels are combined under a single severity level; speech recognition typos; grammatical and/or other language errors (e.g., punctuation); incorrect patient information (e.g., age, sex, medical history, etc.), and/or any other errors. Errors can be corrected, flagged, and/or otherwise addressed). The Campos et al. as modified by Paulett et al. does not disclose generating a third prompt instructing the generative language model to compare the draft custom report with the example financial report.
However, De Ridder discloses generating a third prompt instructing the generative language model to compare the draft custom report with the example financial reports (Paragraph 28 discloses the comparison operation may be iterated as the document is being drafted by the end user).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with De Ridder.’s target content derived modeling and unsupervised language modeling. This would have facilitated post-processing of content (eg a document) for draft improvement.
With respect to claim 11, it is rejected on grounds corresponding to above rejected claim 4, because claim 11 is substantially equivalent to claim 4.
Claim(s) 6 and 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) in further view of Smith et al. (US Patent No. 12124985).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 6, Campos et al. as modified by Paulett et al. does not disclose determining a convergence metric by comparing the custom report with the example financial reports.
However, Smith et al. discloses the method of claim 1, wherein the iterative repeating step comprises:
determining a convergence metric by comparing the custom report with the example financial reports (Column 32 Lines 22-51 discloses compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like); and
continuing the iterative process when the convergence metric indicates insufficient similarity (Column 32 Lines 22-51 discloses compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like); and
terminating the iterative process when the convergence metric exceeds a predetermined threshold indicating substantial replication of the example financial reports (Column 32 Lines 22-51 discloses compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with Smith et al.’s generation of productivity data. This would have facilitated the automated analysis of a data such as a financial report.
With respect to claim 13, it is rejected on grounds corresponding to above rejected claim 6, because claim 13 is substantially equivalent to claim 6.
Claim(s) 7 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) and Mondlock et al. (US Patent No. 12079570).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 1. With respect to claim 7, Campos et al. as modified by Paulett et al. does not disclose receiving user selection of a role-based workflow from a plurality of available workflows, wherein each workflow is associated with different types of financial analysis tasks.
However, Mondlock et al. teaches the method of claim 1, further comprising:
receiving user selection of a role-based workflow from a plurality of available workflows, wherein each workflow is associated with different types of financial analysis tasks (Column 13 Lines 11-25 discloses role, practice, or geographic filters may have been explicitly selected by the user at block 320 and appended to the user query); and
customizing the query processing pipeline based on the selected role-based workflow to apply role-specific filtering rules and data source prioritization (Column Lines discloses RAG pipeline 300 may include at block 350 extracting any role, practice, or geographic filters from the final user query. The filters may be extracted by the document/asset/expert module 122 or any other suitable program).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with Mondlock et al.’s generative AI pipeline. This would have facilitated a customizable answering of a query using a large language model.
With respect to claim 14, it is rejected on grounds corresponding to above rejected claim 7, because claim 14 is substantially equivalent to claim 7.
Claim(s) 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) and Javari et al. (US Pub. No. 20240273128).
The Campos et al. reference as modified by Paulett et al. teaches all the limitations of claim 15. With respect to claim 18, Campos et al. as modified by Paulett et al. does not disclose applying machine learning-based noise reduction algorithms to identify and remove documents that do not match the content requirements of the selected task.
However, Javari et al. teaches the method of claim 15, wherein filtering the retrieved documents to remove irrelevant information comprises:
applying machine learning-based noise reduction algorithms to identify and remove documents that do not match the content requirements of the selected task (Paragraph 85 discloses amount of noise (e.g., number of documents replaced, added, or removed) may be based on an accuracy of the machine-learning algorithm, with more noise being added to a less accurate model to more drastically train the algorithm at early stages of training);
scoring each document based on relevance to the custom query and the selected task (Paragraph 128 discloses determining a relevance value based on the first and second embeddings sets. and in response to the relevance value exceeding a threshold value, presenting the at least one document of the candidate set to the user); and
retaining only documents that exceed a predetermined relevance threshold (Paragraph 128 discloses determining a relevance value based on the first and second embeddings sets. and in response to the relevance value exceeding a threshold value, presenting the at least one document of the candidate set to the user).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting with Javari et al.’s document category recommendations. This would have facilitated presenting relevant document information based on criteria such as context.
Claim(s) 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Campos et al. (US Patent No. 12067039) and Paulett et al. (US Pub. No. 20240347156) and Tomkins et al. (US Pub. No. 20210295822) in further view of Smith et al. (US Patent No. 12124985).
The Campos et al. reference as modified by Paulett et al. and Tomkins et al. teaches all the limitations of claim 15. With respect to claim 20, Campos et al. discloses the method of claim 15, wherein the iterative repeating step comprises:
processing the one or more additional queries through the query processing pipeline to obtain additional query results (Column 10 Lines 50-67 and Column 11 Lines 1-4 discloses a convolutional neural network may have been trained and used to classify (pixelated) image data as characters, photographs, diagrams, media content, and/or other types of information. In some implementations, the extracted information may have been extracted by a document analysis process that uses a pipeline of steps for object detection, object recognition, and/or object classification).
Paulett et al. discloses generating a fourth prompt for the generative language model to synthesize the additional query results with previously obtained query results to generate an updated custom report (Paragraph 95 discloses generating outputs (e.g., text, data, etc.) based on the set of inputs (e.g., unstructured inputs), which can include expanding the information contained within the inputs (e.g., determining a full form version of a shorthand for a finding entered by a radiologist, adding context for clarification, etc.), modifying the format (e.g., language, language style, data format, etc.) of the inputs, and/or otherwise generating outputs).
Campos et al. as modified by Paulett et al. and Tomkins et al. does not disclose comparing the updated custom report with the example financial reports to determine a convergence metric.
However, Smith et al. discloses comparing the updated custom report with the example financial reports to determine a convergence metric (Column 32 Lines 22-51 discloses compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like); and
terminating the iterative process when the convergence metric indicates that the updated custom report substantially matches the format and content structure of the example financial reports within a predetermined threshold (Column 32 Lines 22-51 discloses compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like).
Therefore, it would have been obvious before the effective filing data of invention was made to a person having ordinary skill in the art to modify Campos et al.’s extraction information from documents via a large language model and Paulett et al.’s radiology reporting and Tomkins et al.’s natural language model generation with Smith et al.’s generation of productivity data. This would have facilitated the automated analysis of data such as a financial report.
Relevant Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US PG-Pub. No. 20240346086 is directed to SELF-ORGANIZING MODELING FOR TEXT DATA: [0030] automatically discover one or more optimal category-sets in fully unsupervised fashion without any a priori knowledge from text and/or non-text datasets, and assigns text-based labels to the discovered categories for easier understanding of the discovered categories. This may be applicable to any dataset whose document-meanings can be represented as embeddings or vectors of numbers. It may leverage sentence embeddings of Large Language Models (LLM), or vectors from more conventional character-token-based vectorizers, for example, Tf-IDF (Term Frequency Inverse Document Frequency of records) vectorizers.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICHOLAS E ALLEN whose telephone number is (571)270-3562. The examiner can normally be reached Monday through Thursday 830-630.
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, Boris Gorney can be reached at (571) 270-5626. 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.
/N.E.A/Examiner, Art Unit 2154
/BORIS GORNEY/Supervisory Patent Examiner, Art Unit 2154
Prosecution Insights