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 .
This action is in response to the arguments filed on 08/25/2025. Claims 1-20 are pending in the application and have been considered below.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1:
For Step 1, the claim is a method so it does recite a statutory category of invention.
For Step 2, Prong 1:
The claim recites the limitation of “i) transforming (i.e., determining) an unstructured data in the plurality of construction project documents into structured dataset, or generating a searchable description for an unsearchable content in the plurality of construction project documents.” The transforming or generating limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the transforming or generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of “ii) applying an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents, wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project” The applying limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the applying steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of” querying one or more LLMs, to identify, using one or more LLMs, one or more relevant documents from the plurality of construction project documents and one or more relevant chunks related to the question.” The querying and identifying limitations, as drafted, are processes that, under its broadest reasonable interpretation, cover performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the identifying and generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of “generating a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents.” The generating limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
For Step 2, Prong 2, the claim recites additional elements: “receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents,” and “providing the answer and one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface.”
The “accessing the embeddings generated in (a) and message interface” is a generic computer component to apply an abstract idea under 2106.05(f).
The recited “receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
The “delivering (i.e. transmitting) to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Step 2B
The additional elements of:” accessing the embeddings generated in (a) and “message interface” do not amount to significantly more for the reasons set forth in step 2A above.
Additionally, under the Subject Matter Eligibility (SME), a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “receiving a question from a user via a message interface, wherein the question is related to information from a plurality of construction project documents” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
Here the “delivering (i.e. transmitting) to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of” accessing the embeddings generated in (a) and “message interface” to perform the claim steps amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claim is not patent eligible.
Regarding Claim 2:
Claim 2, which incorporates the rejection of claim 1, recites further limitations such as “
processing (i.e., transforming, determining) the plurality of construction project documents into a plurality of chunks” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 3:
Claim 3, which incorporates the rejection of claim 2, recites further limitations such as “
the one or more relevant documents and the one or more relevant chunks are identified based on dense embeddings and sparse embeddings of the question and the plurality of chunks of the plurality of construction project documents” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 4:
Claim 4, which incorporates the rejection of claim 3, recites further limitations such as “applying a parsing algorithm to generate dense embeddings and sparse embeddings for the plurality of construction project documents and the plurality of chunks” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 5:
Claim 5, which incorporates the rejection of claim 2, recites further limitations such as “
“processing (i.e., transforming, determining) the plurality of construction project documents into the plurality of chunks comprises using one or more chunking techniques” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 6:
Claim 6, which incorporates the rejection of claim 1, recites further limitations such as “
“one or more relevant chunks comprise a word, multiple word, a sentence, a section, a passage, a paragraph, a clause, or a sub-clause” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 7:
Claim 7, which incorporates the rejection of claim 1, recites an additional element:
“allowing the user to access the one or more relevant documents with one or more chunks of texts annotated with a visual indicator.”
The “allowing the user to access the one or more relevant documents with one or more chunks of texts annotated with a visual indicator” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “allowing the user to access (i.e. retrieving) the one or more relevant documents with one or more chunks of texts annotated with a visual indicator” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data.”
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 8:
Claim 8, which incorporates the rejection of claim 1, recites further limitations such as “generating one or more follow-up questions to clarify the question” that are part of the abstract idea.
The claim recites an additional element: “delivering the one or more follow-up questions to the user via the message interface.”
The “delivering the one or more follow-up questions to the user via the message interface” step is a form of insignificant extra-solution activity.” MPEP 2106.05(g).
Here the “delivering (sending or transmitting data) the one or more follow-up questions to the user via the message interface” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data.”
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 9:
Claim 9, which incorporates the rejection of claim 1, recites further limitations such as “generating a summary for each of the plurality of construction project documents” that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 10:
Claim 10, which incorporates the rejection of claim 1, recites further limitations such as “
unstructured or semi- structured document text “that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 11:
Claim 11, which incorporates the rejection of claim 10, recites further limitations such as “unstructured or semi- structured document text comprise at least an image, a table, a spreadsheet, or a combination thereof “that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 12:
Claim 12, which incorporates the rejection of claim 11, recites further limitations such as “generating a searchable description for an unsearchable content in the plurality of construction project documents comprises contextualizing the image, the table, or the spreadsheet based at least in part on metadata to generate a description for the image, the table, or the spreadsheet “that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 13:
Claim 13, which incorporates the rejection of claim 12, recites further limitations such as
“the summary for each of the plurality of construction project documents and the description for the image, the table, or the spreadsheet is searchable “that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 14:
Claim 14, which incorporates the rejection of claim 13, recites an additional element: recites an additional element:
“the one or more links direct the user to an interface and wherein the interface displays the answer and a list of citations used to generate the answer.”
The “one or more links direct the user to an interface and wherein the interface displays the answer and a list of citations used to generate the answer” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “one or more links direct the user to an interface and wherein the interface displays (i.e. sending or transmitting data) the answer and a list of citations used to generate the answer” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data.”
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 15
Claim 15, which incorporates the rejection of claim 1, recites an additional element: “the one or more links direct (i.e. sending) the user to an interface and wherein the interface displays the answer and a list of citations used to generate the answer.”
The “one or more links direct the user to an interface and wherein the interface displays (i.e. sending) the answer and a list of citations used to generate the answer” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Here the “one or more links direct the user to an interface and wherein the interface displays (i.e. sending) the answer and a list of citations used to generate the answer” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 16:
Claim 16, which incorporates the rejection of claim 15, recites an additional element: “the list of citations comprise a list of source documents or a list of relevant passages or sections used to generate the answer “step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the list of citations comprise a list of source documents or a list of relevant passages or sections used to generate the answer “step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 17:
Claim 17, which incorporates the rejection of claim 1, recites further limitations such as
“highlighting the one or more relevant chunks from a linked relevant document used to answer the question “that are part of the abstract idea.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 18
Claim 18, which incorporates the rejection of claim 17, recites an additional element: recites additional elements “clicking on a section of the answer displays the one or
more relevant chunks used to generate the section of the answer.”
The “clicking on a section of the answer displays (i.e. sending or transmitting data) the one or more relevant chunks used to generate the section of the answer” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “clicking on a section of the answer displays (i.e. sending or transmitting data) the one or more relevant chunks used to generate the section of the answer” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 19
Claim 19, which incorporates the rejection of claim 1, recites an additional element: “the LLM prompt includes filenames of the plurality of construction project documents, the question, and a query instruction asking the LLM to select all documents that may be relevant to search for the answer to the user question.”
The “LLM prompt includes filenames of the plurality of construction project documents, the question, and a query instruction asking the LLM to select all documents that may be relevant to search for the answer to the user question” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “LLM prompt includes filenames of the plurality of construction project documents, the question, and a query instruction asking the LLM to select (i.e. retrieving).all documents that may be relevant to search for the answer to the user question” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
There are no additional elements recited in this claim that amount to an integration of the judicial exception into a practical application or significantly more than the judicial exception. Therefore, the claim is not eligible.
Regarding Claim 20:
For Step 1, the claim is a computer implemented system so it does recite a statutory category of invention.
For Step 2, Prong 1:
The claim recites the limitation of “generating embeddings of a plurality of construction project documents.” The generating limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the generating steps from practically being performed in the human mind. This limitation is a mental process.
The claim recites the limitation of “i) transforming (i.e., determining) an unstructured data in the plurality of construction project documents into structured dataset, or generating a searchable description for an unsearchable content in the plurality of construction project documents.” The transforming or generating limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the transforming or generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of “ii) applying an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents, wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project” The applying limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the applying steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of “generating a large language model (LLM) prompt based at least in part on the question.” The generating limitations, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of” querying one or more LLMs, to identify, using one or more LLMs, one or more relevant documents from the plurality of construction project documents and one or more relevant chunks related to the question.” The querying, and identifying limitations, as drafted, are processes that, under its broadest reasonable interpretation, cover performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the identifying and generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
The claim recites the limitation of “generating a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents.” The generating limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is nothing in the claim precludes the generating steps from practically being performed in the human mind. This limitation is a mental process.
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process”
For Step 2, Prong 2, the claim recites additional elements: processor, memory, : “receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents.” and “providing the answer and one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface.”
The processor is recited at a high level of generality, i.e., as a generic processor performing a generic computer function of processing data. This generic processor limitation is no more than mere instructions to apply the exception using a generic computer component. (MPEP 2106.05(f)).
The “accessing the embeddings generated in (a) and message interface” is a generic computer component to apply an abstract idea under 2106.05(f).
The recited “receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
The “delivering (i.e. transmitting) to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface” step is a form of insignificant extra-solution activity. See MPEP 2106.05(g).
Step 2B
The additional elements of “processor, memory, accessing the embeddings generated in (a) and “message interface” do not amount to significantly more for the reasons set forth in step 2A above.
Additionally, under the Subject Matter Eligibility (SME), a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be reevaluated in Step 2B.
Here the “receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
Here the “delivering (i.e. transmitting) to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface” step was considered to be extra-solution activity in Step 2A, and thus it is reevaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The addition of insignificant extra-solution activity does not amount to an inventive concept, particularly when the activity is well-understood or conventional (MPEP 2106.05(d)). This appears to be well-understood, routine, conventional as evidenced by MPEP 2106.05(d)(II)(iv).
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data”.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of “processor, memory, accessing the embeddings generated in (a) and “message interface” to perform the claim steps amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2, 5-6 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER).
As to claim 1, Buliani teaches a method for generative question answering for construction project documents, the method comprising:
(a) generating embeddings of a plurality of construction project documents at least by:
i) transforming an unstructured data in the plurality of construction project documents into structured dataset, or generating a searchable description for an unsearchable content in the plurality of construction project documents, and
ii) applying an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents,
wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project (col. 14, lines 12-51, Such documentation can include software project descriptions, software documentation, source code files, ticketing systems, and the like from other public software programs or systems. Another service 194 of the provider network 100 Augmented Generation (RAG) techniques to identify documents of "relevance" to a given task. Initially, each of the available documents with the documentation service can be encoded as an embedding, those embeddings stored in a database. When invoked, the documentation aggregator can 223 use the encoder to generate an embedding from user text may be a documentation services that stores such documentation. A documentation aggregator 223 can use Retrieval Augmented Generation (RAG) techniques to identify documents of "relevance" to a given task. Initially, each of the available documents with the documentation service can be encoded as an embedding, those embeddings stored in a database. When invoked, the documentation aggregator can 223 use the encoder to generate an embedding from user text for the given task. The document aggregator 223 can then identify relevant documents based on the distance between the task embedding and document embeddings in the database, selecting the N nearest document embeddings, document embeddings within some distance threshold, or some other criteria to identify documents having embeddings in proximity to the task embedding. The documentation aggregator 223 can then access the documentation service to obtain the documents associated with those selected embeddings).
receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents (Fig 4, EXAMPLE PROMPT TEMPLATES 412 “YOU CAN ASK CLARIFYING QUESTIONS ABOUT THE PROJECT. LIMIT THE NUMBER OF CLARIFYING QUESTIONS TO NO MORE THAN TEN QUESTIONS. RESPOND {RESPONSE_DEF NUMBERED_QUESTIONS_NOCONTEXT});
generating a large language model (LLM) prompt based at least in part on the question (Fig 4, EXAMPLE PROMPT TEMPLATES 414 “RESPONSE_DEF _ YN="YES OR NO"; Fig.5, elements 214, 502, 504, 506, 510, 512, 514, 516 and 599A; col 16 lines 59-67 to col. 17, lines 1-35 In some examples, given a response indicating a previous response did not comply with a response definition, the validation agent can prompt the queried LLM to provide an explanation, which can be included in the response correction prompt (e.g., at operation 910). At operation 512, the validation agent receives an updated prompt response from the LLM 599A. At operation 514, the validation agent sends a prompt like the one of that of operation 506);
querying one or more LLMs, one or more to identify relevant documents from the plurality of construction project documents and [one or more relevant chunks] related to the question by accessing the embeddings generated in (a) (col. 14, lines 12-51, a documentation aggregator 223 can use Retrieval Augmented Generation (RAG) techniques to identify documents of "relevance" to a given task. Initially, each of the available documents with the documentation service can be encoded as an embedding, those embeddings stored in a database. When invoked, the documentation aggregator can 223 use the encoder to generate an embedding from user text for the given task. The document aggregator 223 can then identify relevant documents based on the distance between the task embedding and document embeddings in the database, selecting the N nearest document embeddings, document embeddings within some distance threshold, or some other criteria to identify documents having embeddings in proximity to the task embedding. The documentation aggregator 223 can then access the documentation service to obtain the documents associated with those selected embeddings; col. 17, lines 1-35 In some examples, given a response indicating a previous response did not comply with a response definition, the validation agent can prompt the queried LLM to provide an explanation, which can be included in the response correction prompt (e.g., at operation 910). At operation 512, the validation agent receives an updated prompt response from the LLM 599A. At operation 514, the validation agent sends a prompt like the one of that of operation 506).
However, Buliani teaches context data but fails to explicitly teach:
d)querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question by accessing the embeddings generated in (a)
(e) generating a context data by recursively summarizing [the identified
one or more relevant chunks across the one or more documents], and
generating an answer based at least in part on the context data and the LLM prompt;
(f) delivering to the user, i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface
Vasylyev, in combination with Buliani, teaches:
(e) generating a context data by recursively summarizing [the identified
one or more relevant chunks across the one or more documents] (paragraphs [0307] –[0308], efficiently manage extensive data and long conversation histories by summarizing it in a recursive manner and store it in contextual memory unit 116… ensure an even more efficient utilization of contextual memory unit 116, the recursive summaries may be stored in RAM 124 and retrieved only in cases when the GPT model of assistant system 2 determines that access to those summaries is needed in the context of the ongoing conversation, for example.), and
generating an answer based at least in part on the context data and the LLM prompt (paragraph [0283] For example, after the transformer-based language model processes the voice input and generates a conversational response, the system tokenizes the generated response using techniques such as word-level, subword-level, or character-level tokenization. The tokenized representation of the generated response is then stored in contextual memory unit 116, along with the tokenized representation of
the voice input and any other relevant contextual information. The tokenized representations in contextual memory unit 116 can be used by the transformer-based language model for subsequent processing, such as generating follow-up responses or answering user queries that rely on the conversation history. The sliding window mechanism employed by contextual memory unit 116 ensures that the most recent and relevant parts of the conversation are readily accessible for processing. In alternative embodiments, the generated conversational response may be stored in contextual
memory unit 116 in other formats, such as raw text, embeddings, or compressed representations, depending on the specific requirements and design choices…).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of Buliani to add d recursive summaries to the system of Buliani as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to efficiently manage extensive data and long conversation histories by summarizing it in a recursive manner, as suggested by Vasylyev, ([0307]).
However, Buliani and Vasylyev fail to explicitly teach:
d) querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question [by accessing the embeddings generated in (a)];
the identified one or more relevant chunks across the one or more documents;
f) delivering to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface.
GRENADER, in combination with Buliani and Vasylyev, teaches:
d) querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question [by accessing the embeddings generated in (a)] ([0081]-[0083] This results in query results 218. In some cases, the query results might be complex or compound so, accordingly, step 222 might be accomplished by performing several iterations of asking the LLM to reformat the query results (e.g., reformatted query results 227) into a large language model system answer (e.g., LLM answer 226) that comports with the requested reformatting (e.g., into the natural language of CMS user 101); [0154] and [0159] As a result, and responsive to performance of the evaluations within chunk selection module 370, one or more instances of selected chunks 375 are provided to prompt generation module 374; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter);
[generating a context data by recursively summarizing] the identified one or more relevant chunks across the one or more documents ([0154] and [0159] As a result, and responsive to performance of the evaluations within chunk selection module 370, one or more instances of selected chunks 375 are provided to prompt generation module 374; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter);and
f) delivering to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface (paragraphs [0066]- [0069], there are also many situations where a user question can be best answered using external LLM resources (e.g., computing and storage resources of a selected LLM). In those situations where a user question can be best answered using external LLM resources, an LLM agent (Examiner interprets LLM agent as a link) is deployed to interact with the external LLM resources; [0159]-]0160] As shown, performance of the operations within prompt generation module 374 result in one or more LLM prompts 327, which in turn are provided to a selected LLM (not shown) with the expectation that the LLM will return a large language model system answer (e.g., LLM answer 226, as shown) corresponding to the user question; ; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani and Vasylyev to add links to the relevant chunks to the combination system of Buliani and Vasylyev as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to select the relevant chunks, as suggested by GRENADER, ([0171]).
As to claim 2, which incorporates the rejection of claim 1, Buliani and Vasylyev fail to explicitly teach: processing the plurality of construction project documents into a plurality of chunks
GRENADER teaches: processing the plurality of construction project documents into a plurality of chunks (paragraphs [0121]- [0123] As shown, the chunk selection technique 3A500 takes as input, a user question embedding 320 and produces both selected-in chunks (e.g., closest matches 326) as well as context 323. The selected-in chunks, together with its context, can then be used in a prompt that is in tum provided to an LLM).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani and Vasylyev to add links to the relevant chunks to the combination system of Buliani and Vasylyev as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to select the relevant chunks, as suggested by GRENADER, ([0171]).
As to claim 5, which incorporates the rejection of claim 2, Buliani and Vasylyev fail to explicitly teach wherein processing the plurality of construction project documents into the plurality of chunks comprises using one or more chunking techniques.
GRENADER discloses:
wherein processing the plurality of construction project documents into the plurality of chunks comprises using one or more chunking techniques (paragraphs [0029]-[0030] FIG. 3A5 shows a chunk selection technique, according to an embodiment).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani and Vasylyev to add links to the relevant chunks to the combination system of Buliani and Vasylyev as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to select the relevant chunks, as suggested by GRENADER, ([0171]).
As to claim 6, which incorporates the rejection of claim 1, Buliani and Vasylyev fail to explicitly teach wherein the one or more relevant chunks comprise a word, multiple word, a sentence, a section, a passage, a paragraph, a clause, or a sub-clause.
GRENADER discloses wherein the one or more relevant chunks comprise a word, multiple word, a sentence, a section, a passage, a paragraph, a clause, or a sub-clause (paragraph [0031] FIG. 3A6 is a diagram that shows an example chunk type assignment technique where individual portions of a document or documents are scored for relevance with respect to a provided user question, according to an embodiment).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani and Vasylyev to add links to the relevant chunks to the combination system of Buliani and Vasylyev as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to select the relevant chunks, as suggested by GRENADER, ([0171]).
As to claim 20, Buliani teaches a computer implemented system for generative question answering for construction project documents, the system comprising at least one processor,
a memory, and instructions executable by the at least one processor to cause the at least one processor to perform (Fig. 303, element 3020, memory and element 3010A, processor):
(a) generating embeddings of a plurality of construction project documents at least by:
i) transforming an unstructured data in the plurality of construction project documents into structured dataset, or generating a searchable description for an unsearchable content in the plurality of construction project documents, and
ii) applying an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents,
wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project (col. 14, lines 12-51 Such documentation can include software project descriptions, software documentation, source code files, ticketing systems, and the like from other public software programs or systems. Another service 194 of the provider network 100 Augmented Generation (RAG) techniques to identify documents of "relevance" to a given task. Initially, each of the available documents with the documentation service can be encoded as an embedding, those embeddings stored in a database. When invoked, the documentation aggregator can 223 use the encoder to generate an embedding from user text may be a documentation services that stores such documentation. A documentation aggregator 223 can use Retrieval Augmented Generation (RAG) techniques to identify documents of "relevance" to a given task. Initially, each of the available documents with the documentation service can be encoded as an embedding, those embeddings stored in a database. When invoked, the documentation aggregator can 223 use the encoder to generate an embedding from user text for the given task. The document aggregator 223 can then identify relevant documents based on the distance between the task embedding and document embeddings in the database, selecting the N nearest document embeddings, document embeddings within some distance threshold, or some other criteria to identify documents having embeddings in proximity to the task embedding. The documentation aggregator 223 can then access the documentation service to obtain the documents associated with those selected embeddings).
b) receiving a question from a user via a message interface, wherein the question is related to querying information from a plurality of construction project documents (Fig 4, EXAMPLE PROMPT TEMPLATES 412 “YOU CAN ASK CLARIFYING QUESTIONS ABOUT THE PROJECT. LIMIT THE NUMBER OF CLARIFYING QUESTIONS TO NO MORE THAN TEN QUESTIONS. RESPOND {RESPONSE_DEF NUMBERED_QUESTIONS_NOCONTEXT});
c) generating a large language model (LLM) prompt based at least in part on the question (Fig 4, EXAMPLE PROMPT TEMPLATES 414 “RESPONSE_DEF _ YN="YES OR NO"; Fig.5, elements 214, 502, 504, 506, 510, 512, 514, 516 and 599A; col 16 lines 59-67 to col. 17, lines 1-35 In some examples, given a response indicating a previous response did not comply with a response definition, the validation agent can prompt the queried LLM to provide an explanation, which can be included in the response correction prompt (e.g., at operation 910). At operation 512, the validation agent receives an updated prompt response from the LLM 599A. At operation 514, the validation agent sends a prompt like the one of that of operation 506);
d) querying one or more LLMs, one or more to identify relevant documents from the plurality of construction project documents and one or more relevant chunks related to the question by accessing the embeddings generated in (a) (col. 17, lines 1-35 In some examples, given a response indicating a previous response did not comply with a response definition, the validation agent can prompt the queried LLM to provide an explanation, which can be included in the response correction prompt (e.g., at operation 910). At operation 512, the validation agent receives an updated prompt response from the LLM 599A. At operation 514, the validation agent sends a prompt like the one of that of operation 506).
However, Buliani teaches context data but fails to explicitly teach:
d)querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question by accessing the embeddings generated in (a)
(e) generating a context data by recursively summarizing [the identified
one or more relevant chunks across the one or more documents], and
generating an answer based at least in part on the context data and the LLM prompt;
(f) delivering to the user, i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface
Vasylyev, in combination with Buliani, teaches:
(e) generating a context data by recursively summarizing [the identified
one or more relevant chunks across the one or more documents] (paragraphs [0307] –[0308], efficiently manage extensive data and long conversation histories by summarizing it in a recursive manner and store it in contextual memory unit 116… ensure an even more efficient utilization of contextual memory unit 116, the recursive summaries may be stored in RAM 124 and retrieved only in cases when the GPT model of assistant system 2 determines that access to those summaries is needed in the context of the ongoing conversation, for example.), and
generating an answer based at least in part on the context data and the LLM prompt (paragraph [0283] For example, after the transformer-based language model processes the voice input and generates a conversational response, the system tokenizes the generated response using techniques such as word-level, subword-level, or character-level tokenization. The tokenized representation of the generated response is then stored in contextual memory unit 116, along with the tokenized representation of
the voice input and any other relevant contextual information. The tokenized representations in contextual memory unit 116 can be used by the transformer-based language model for subsequent processing, such as generating follow-up responses or answering user queries that rely on the conversation history. The sliding window mechanism employed by contextual memory unit 116 ensures that the most recent and relevant parts of the conversation are readily accessible for processing. In alternative embodiments, the generated conversational response may be stored in contextual
memory unit 116 in other formats, such as raw text, embeddings, or compressed representations, depending on the specific requirements and design choices…).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the system of Buliani to add d recursive summaries to the system of Buliani as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to efficiently manage extensive data and long conversation histories by summarizing it in a recursive manner, as suggested by Vasylyev, ([0307]).
However, Buliani and Vasylyev fail to explicitly teach:
d) querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question [by accessing the embeddings generated in (a)];
the identified one or more relevant chunks across the one or more documents;
f) delivering to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface.
GRENADER, in combination with Buliani and Vasylyev, teaches:
d) querying one or more LLMs, one or more to identify [relevant documents from the plurality of construction project documents and] one or more relevant chunks related to the question [by accessing the embeddings generated in (a)] ([0081]-[0083] This results in query results 218. In some cases, the query results might be complex or compound so, accordingly, step 222 might be accomplished by performing several iterations of asking the LLM to reformat the query results (e.g., reformatted query results 227) into a large language model system answer (e.g., LLM answer 226) that comports with the requested reformatting (e.g., into the natural language of CMS user 101); [0154] and [0159] As a result, and responsive to performance of the evaluations within chunk selection module 370, one or more instances of selected chunks 375 are provided to prompt generation module 374; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter);
[generating a context data by recursively summarizing] the identified one or more relevant chunks across the one or more documents ([0154] and [0159] As a result, and responsive to performance of the evaluations within chunk selection module 370, one or more instances of selected chunks 375 are provided to prompt generation module 374; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter);and
f) delivering to the user i) the answer and ii) one or more links to the one or more relevant chunks used to generate the answer in a text message in the message interface (paragraphs [0066]- [0069], there are also many situations where a user question can be best answered using external LLM resources (e.g., computing and storage resources of a selected LLM). In those situations where a user question can be best answered using external LLM resources, an LLM agent (Examiner interprets LLM agent as a link) is deployed to interact with the external LLM resources; [0159]-]0160] As shown, performance of the operations within prompt generation module 374 result in one or more LLM prompts 327, which in turn are provided to a selected LLM (not shown) with the expectation that the LLM will return a large language model system answer (e.g., LLM answer 226, as shown) corresponding to the user question; ; [0171]-[0172], then relevant chunks (e.g., based at least in part on historical interactions) are selected thereafter).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani and Vasylyev to add links to the relevant chunks to the combination system of Buliani and Vasylyev as taught by Vasylyev, above. The modification would have been obvious because one of ordinary skill would be motivated to select the relevant chunks, as suggested by GRENADER, ([0171]).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and LASSANCE et al. (US 2023/0021996 A1, hereinafter referred to as LASSANCE).
As to claim 3, which incorporates the rejection of claim 2, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the one or more relevant documents and the one or more relevant chunks are identified based on dense embeddings and sparse embeddings of the question and the plurality of chunks of the plurality of construction project documents.
However, LASSANCE, in combination with Buliani, Vasylyev and GRENADER, teaches wherein the one or more relevant documents and the one or more relevant chunks are identified based on dense embeddings and sparse embeddings of the question and the plurality of chunks of the plurality of construction project documents (paragraphs [0041], chunks; [0064]-[0066], dense embeddings and sparse embeddings).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add dense embeddings and sparse embeddings to the combination system of Buliani, Vasylyev and GRENADER, as taught by LASSANCE, above. The modification would have been obvious because one of ordinary skill would be motivated to generate an inverted index for further improvement to retrieval efficiency, as suggested by LASSANCE, ([0033]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and LASSANCE et al. (US 2023/0021996 A1, hereinafter referred to as LASSANCE), and further in view of DUAN et al. (US 2023/0359441 A1, hereinafter referred to as DUAN).
As to claim 4, which incorporates the rejection of claim 3, Buliani, Vasylyev, GRENADER and LASSANCE fail to explicit teach wherein processing the plurality of construction project documents comprises applying a parsing algorithm to generate dense embeddings and sparse embeddings for the plurality of construction project documents and the plurality of chunks.
However, DUAN, in combination with Buliani, Vasylyev, GRENADER and LASSANCE, teaches wherein processing the plurality of construction project documents comprises applying a parsing algorithm to generate dense embeddings and sparse embeddings for the plurality of construction project documents and the plurality of chunks (paragraphs [0035]- [0036], wherein using the broadest reasonable interpretation, Examiner interprets the combination of elements 208 (i.e., parser) and element 122 (i.e., retrieval source code database) to teach the limitation).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev, GRENADER and LASSANCE to add a parsing algorithm to the combination system of Buliani, Vasylyev, GRENADER and LASSANCE, as taught by DUAN, above. The modification would have been obvious because one of ordinary skill would be motivated to generate an inverted index for further improvement to retrieval efficiency, as suggested by LASSANCE, ([0033]).
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and Salvati (US 2020/0320141 A1, hereinafter referred to as Salvati).
As to claim 7, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the one or more links direct the user to a webpage allowing the user to access the one or more relevant documents with one or more chunks of texts annotated with a visual indicator.
However, Salvati, in combination with Buliani, Vasylyev and GRENADER, teaches wherein the one or more links direct the user to a webpage allowing the user to access the one or more relevant documents with one or more chunks of texts annotated with a visual indicator (paragraph [0037] In some instances, the user interface may be rendered and accessed
in a browser in which the pages load up, viewed by a user, and the user may highlight, annotate, and export the information).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add a webpage to the combination system of Buliani, Vasylyev and GRENADER, as taught by Salvati, above. The modification would have been obvious because one of ordinary skill would be motivated to provide a more responsive environment rather than waiting for the original file to load to present the summary records, as suggested by Salvati, ([0037]).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and Poddar et al. (US 2021/0118442 A1, hereinafter referred to as Poddar).
As to claim 8, which incorporates the rejection of claim 2, Buliani, Vasylyev and GRENADER fail to explicit teach:
generating one or more follow-up questions to clarify the question and delivering the one or more follow-up questions to the user via the message interface.
However, Poddar, in combination with GRENADER, teaches:
generating one or more follow-up questions to clarify the question and delivering the one or more follow-up questions to the user via the message interface
(paragraph [0155] For instance, after the user 1114 receives the response (e.g., "Your keys are on the kitchen counter.") to the initial request, she may then ask a follow-up question involving a further relationship between the keys and other objects (for example, "What are they next to?"). As an example and not by way of limitation, as
discussed above, the assistant system 140 may perform coreference resolution and determine that the coreference "they" is still referring to the keys 1120. As another example and not by way of limitation, the assistant system 140 may perform slot tracking or slot carry over, in which item mentions associated with previous utterances are carried over, by determining that the keys 1120 are still a relevant target reference associated with the current utterance without specifically resolving the target for the word "they". In another particular embodiment in which the requested relational information was already determined in the first pass, the assistant system 140 may access the knowledge graph 1140 to determine the requested relational information and return this relational information in response to the user request (e.g., "Your keys are next to the plant").
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add one or more follow-up questions to the combination system of Buliani, Vasylyev and GRENADER as taught by Poddar, above. The modification would have been obvious because one of ordinary skill would be motivated to generate a requested relational information pertaining to a query, as suggested by Poddar, ([0155]).
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and SOMECH et al. (US 2024/0419912 A1, hereinafter referred to as SOMECH).
As to claim 9, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach: generating a summary for each of the plurality of construction project documents.
However, SOMECH, in combination with Buliani, Vasylyev and GRENADER fail, teaches: generating a summary for each of the plurality of construction project documents (paragraphs [0004], LLM will generate a summary of the meeting in a format like the example; [0027] Responsively, the model generates a first output that comprises: (1) the meeting summary; [0032] However, the model may inadvertently copy a phone number from an example it used to generate the summary and inadvertently incorporate the phone number in the generated summary of document D).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add a document summary to the combination system of Buliani, Vasylyev and GRENADER as taught by SOMECH, above. The modification would have been obvious because one of ordinary skill would be motivated to increase data security or privacy, as suggested by SOMECH, ([0032]).
Claims 10-12 are rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and TUNSTALL-PEDOE et al. (US 2023/0259705 A1, hereinafter referred to as TUNSTALL-PEDOE).
As to claim 10, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the unstructured data in the plurality
of construction project document comprise unstructured or semi-structured document text.
However, TUNSTALL-PEDOE, in combination with GRENADER, teaches wherein the unstructured data in the plurality of construction project document comprise unstructured or semi- structured document text (paragraphs [0027]-[0032] providing a structured representation of data and a system operable to process language with reference to the structured representation of data to provide additional useful text relevant to the language; [0457] Structured data).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add unstructured or semi- structured document text to the combination system of Buliani, Vasylyev and GRENADER as taught by TUNSTALL-PEDOE, above. The modification would have been obvious because one of ordinary skill would be motivated to assign semantic nodes to the identifiers in the structured data or relations corresponding to the relations in a relational database, as suggested by TUNSTALL-PEDOE ([0457]).
As to claim 11, which incorporates the rejection of claim 10, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the unstructured or semi- structured document text comprise at least an image, a table, a spreadsheet, or a combination thereof.
However, TUNSTALL-PEDOE, in combination with GRENADER, teaches wherein the unstructured or semi- structured document text comprise at least an image, a table, a spreadsheet, or a combination thereof (paragraph [0457] Structured data such as the content of a table found in a document or on the web, a spreadsheet or the contents of a relational, graph or other database).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add unstructured or semi- structured document text to the combination system of Buliani, Vasylyev and GRENADER as taught by TUNSTALL-PEDOE, above. The modification would have been obvious because one of ordinary skill would be motivated to assign semantic nodes to the identifiers in the structured data or relations corresponding to the relations in a relational database, as suggested by TUNSTALL-PEDOE ([0457]).
As to claim 12, which incorporates the rejection of claim 11, Buliani, Vasylyev and GRENADER fail to explicit teach wherein generating a searchable description
for an unsearchable content in the plurality of construction project documents comprises contextualizing the image, the table, or the spreadsheet based at least in part on metadata to generate a description for the image, the table, or the spreadsheet.
However, TUNSTALL-PEDOE, in combination with Buliani, Vasylyev and GRENADER, teaches wherein generating a searchable description
for an unsearchable content in the plurality of construction project documents comprises contextualizing the image, the table, or the spreadsheet based at least in part on metadata to generate a description for the image, the table, or the spreadsheet (paragraphs [0022], provide new context data for the LLM, in order to improve the
output, such as continuation text output, generated by the LLM in response to a prompt; [0235] Context can be used to provide nuance, source, belief, temporal validity
etc.; 0457] Structured data such as the content of a table found in a document or on the
web, a spreadsheet or the contents of a relational, graph or other database; [1601] and [2157], structured data, such as the content of a table found in a document or on the web, a spreadsheet or the content of a relational, graph or other database is turned
into semantic nodes and passages by assigning semantic nodes to the identifiers in the structured data and writing semantic nodes and passages that correspond to the meaning of the structured data).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add context to the combination system of Buliani, Vasylyev and GRENADER as taught by TUNSTALL-PEDOE, above. The modification would have been obvious because one of ordinary skill would be motivated to order to improve the output, as suggested by TUNSTALL-PEDOE ([0022]).
Claims 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and TUNSTALL-PEDOE et al. (US 2023/0259705 A1, hereinafter referred to as TUNSTALL-PEDOE), and Salvati (US 2020/0320141 A1, hereinafter referred to as Salvati).
As to claim 13, which incorporates the rejection of claim 12, Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE fail to explicit teach wherein the summary for each of the plurality of construction project documents and the description for the image, the table, or the spreadsheet is searchable.
However, Salvati, in combination with Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE, teaches wherein the summary for each of the plurality of construction project documents and the description for the image, the table, or the spreadsheet is searchable (paragraphs [0038]-[0040] The record report may be a summary report, the summary table 110, or any data structure representative of a record of a set of records. In some embodiments, the set of records may be placed at the end of the summary report).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE to add summary to the combination system of Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE, as taught by Salvati, above. The modification would have been obvious because one of ordinary skill would be motivated to enable the user to search, highlight, annotate, and export the summary tables, as suggested by Salvati, ([0026]).
As to claim 14, which incorporates the rejection of claim 13, Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE fail to explicit teach wherein the one or more links are configured to link back to the summary or the description.
However, Salvati, in combination with Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE, teaches wherein the one or more links are configured to link back to the summary or the description (paragraphs [0039]-[0044] The user interface may also jump back to the corresponding hyperlink in the summary report after selecting a backtracking hyperlink 220 associated with the record of the set of records. For example, selecting advancing hyperlink 210 causes the user interface to jump forward to a record in the set of records on pages 65 and 66 with a single click. In another example, selecting backtracking hyperlink 220 causes the user interface to jump back to the corresponding summary table).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE to add links to the combination system of Buliani, Vasylyev, GRENADER and TUNSTALL-PEDOE, as taught by Salvati, above. The modification would have been obvious because one of ordinary skill would be motivated to have hyperlinks across the
record reporting system enables the user interface to quickly identify and present important information, as suggested by Salvati, ([0026]).
Claims 15-16 are rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and Khosla et al. (US 2025/0005058 A1, hereinafter referred to as Khosla).
As to claim 15, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the one or more links direct the user to an interface and wherein the interface displays the answer and a list of citations used to generate the answer.
However, Khosla, in combination with Buliani, Vasylyev and GRENADER, teaches wherein the one or more links direct the user to an interface and wherein the interface displays the answer and a list of citations used to generate the answer (paragraphs [0026] the attribution component 109 may provide links (e.g., uniform resource identifiers (URI)) or titles as references (e.g., links to webpages, links to online documents, links to images, audio, video, etc.) to documents whose passages were retrieved by the aggregator component; [0066]-[0067], the attribution component 109 may provide references to the retrieved passages, inline citations to sentences of retrieved passages used in the answer, or provide similar questions to the natural language question. For example, the attribution component 109 may provide reference links and titles to the retrieved passages used by the LLM component 106 (e.g., retrieved passages used as context to generate the answer), which may allow the submitter of the question to get more details on the referenced passages; the attribution component 109 may determine that a sentence (e.g., or span of words) from the retrieved passages was used directly ( e.g., word for word or verbatim) in an answer generated by the LLM component 106, and cite that retrieved passage (e.g., or document associated with the retrieved passage) inline with the sentence (e.g., or span of words) used in the answer (e.g., the sentence used in the answer will have a citation citing the passage or document where the sentence was taken from); [0076], the answer and retrieved passages are sent to the customer computing device 122.).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add links and citations to the combination system of Buliani, Vasylyev and GRENADER as taught by Khosla, above. The modification would have been obvious because one of ordinary skill would be motivated to provide reference links and titles to the retrieved passages used by the LLM component 106 (e.g., retrieved passages used as context to generate the answer), which may allow the
submitter of the question to get more details on the referenced passages, as suggested by Khosla ([0075]).
As to claim 16, which incorporates the rejection of claim 15, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the list of citations comprises a list of source documents or a list of relevant passages or sections used to generate the answer.
However, Khosla, in combination with Buliani, Vasylyev and GRENADER, teaches wherein the list of citations comprises a list of source documents or a list of relevant passages or sections used to generate the answer (paragraphs [0026], references or other source identifiers to the retrieved passages, inline citations to sentences from retrieved passages in answers, and provide similar questions to the original question posed. For example, the attribution component 109 may provide links (e.g., uniform resource identifiers (URI)) or titles as references (e.g., links to webpages, links to online documents, links to images, audio, video, etc.) to documents whose passages were retrieved by the aggregator component 104).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add links and citations to the combination system of Buliani, Vasylyev and GRENADER as taught by Khosla, above. The modification would have been obvious because one of ordinary skill would be motivated to provide reference links and titles to the retrieved passages used by the LLM component 106 (e.g., retrieved passages used as context to generate the answer), which may allow the submitter of the question to get more details on the referenced passages, as suggested by Khosla ([0075]).
Claims 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and ABRAMS et al. (US 2024/0256841 A1, hereinafter referred to as ABRAMS).
As to claim 17, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach: highlighting the one or more relevant chunks from a linked relevant document used to answer the question.
However, ABRAMS, in combination with Buliani, Vasylyev and GRENADER, teaches: highlighting the one or more relevant chunks from a linked relevant document used to answer the question (paragraph [0059], the generative model 112 is provided with information selected by the user (e.g., when the user highlights a portion of a webpage): (Examiner interprets the portion of a webpage as a relevant chunk). For large documents, techniques to summarize parts of the document to form a document index or techniques to send one or more chunks/snippets of the document into a limited size buffer of the generative model 112 can be utilized. The generative model 112 can output links to other webpages, and such links can cause the web browser 136 to load new pages when selected. User dialogue may be used to navigate within the page or to other pages or may be used to open pages in new tabs or in a current tab. Answers to
questions or passages found for a search/find query can be highlighted in the main document pane, while the side panel may also have access to the rendered image of the main pages to allow conversation, search/find, question and answer over non text elements such as images or figures).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add links and citations to add “highlighting” the combination system of Buliani, Vasylyev and GRENADER as taught by ABRAMS, above. The modification would have been obvious because one of ordinary skill would be motivated to have answers to questions or passages found for a search/find query can be highlighted in the main document pane, as suggested by ABRAMS, ([0059]).
As to claim 18, which incorporates the rejection of claim 17, Buliani, Vasylyev and GRENADER fail to explicit teach wherein clicking on a section of the answer displays the one or more relevant chunks used to generate the section of the answer.
However, ABRAMS, in combination with Buliani, Vasylyev and GRENADER, teaches wherein clicking on a section of the answer displays the one or more relevant chunks used to generate the section of the answer (paragraph [0059] User dialogue may be used to navigate (Examiner interprets “navigate” as clicking) within the page or to other pages or may be used to open pages in new tabs or in a current tab. Answers to questions or passages found for a search/find query can be highlighted in the main document pane, while the side panel may also have access to the rendered image of the main pages to allow conversation, search/find, question and answer over non text elements such as images or figures)..
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add “clicking section” to combination system of Buliani, Vasylyev and GRENADER as taught by ABRAMS, above. The modification would have been obvious because one of ordinary skill would be motivated to have contextual weight given to main page content that is currently scrolled into view and or where the user has spent more time viewing, as suggested by ABRAMS, ([0059]).
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Buliani et al. (US 12,530,173 B1, hereinafter referred to as Buliani), in view of Vasylyev (US 2024/0412720 A1, hereinafter referred to as Vasylyev), and further in view of GRENADER et al. (US 2024/0362476 A1, hereinafter referred to as GRENADER), and Thomas et al. (US 12,147,758 B1, hereinafter referred to as Thomas)
As to claim 19, which incorporates the rejection of claim 1, Buliani, Vasylyev and GRENADER fail to explicit teach wherein the LLM prompt includes filenames of the plurality of construction project documents, the question, and a query instruction asking the LLM to select all documents that may be relevant to search for the answer to the user question.
However, Thomas, in combination with Buliani, Vasylyev and GRENADER, teaches wherein the LLM prompt includes filenames of the plurality of construction project documents, the question, and a query instruction asking the LLM to select all documents that may be relevant to search for the answer to the user question (col. 8, lines 12-21 The computing device generates a prompt for the LLM service based on the input (step 203). In an implementation, the prompt includes spreadsheet metadata, such as column, headers, row headers, a document or filename, a sheet name, and so on; col.10, lines 53-67, a prompt engine of the application service receives table metadata 401 associated with a table of spreadsheet data including row and/or column headers, row and/or column data types, table headers, sheet names, workbook titles, document or filenames, and so on; col. 9, lines 61-66, receiving user input (i.e. query) and displaying response to the input; col. 10, lines 1-23, an LLM model such as a GPT model, is configured by a prompt engine of an application service including metadata input 301, user input 303 and context information 305. In an implementation, metadata input 301 includes data-related information such as row and/or column headers, row and/or column data types, table, headers, sheet names, workbook titles, document or filenames, and so on. User input 303 includes a user query submitted by a user in the task pane of the user interface).
It would have been obvious to one of ordinary skill in the art before the effective filing of
the claimed invention to modify the combination system of Buliani, Vasylyev and GRENADER to add LLM prompt with filenames to the combination system of Buliani, Vasylyev and GRENADER as taught by Thomas, above. The modification would have been obvious because one of ordinary skill would be motivated to balance prompt size with providing sufficient information to generate a useful response, as suggested by Thomas, (col. 5, lines 65-67).
Response to Applicant’s arguments
Applicant's arguments on file on 08/25//2025 with respect to prior art rejection of claims 1-20 have been considered, are moot in vie w of new ground(s) of rejection, and are not persuasive for the 101 rejections.
35 U.S.C. § 101
Argument (page 18)
Applicant respectfully submits that the claims, as amended, are eligible for at least the following reasons.
Examiner's response:
Examiner respectfully disagrees. Applicant appears to assert that the amended claim is subject matter eligible.
Applicant respectfully submits that the claims, as amended, are eligible for at least the following reasons.
MPEP 2106.04(a) “Mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion).” See additionally see MPEP 2106.04(a)(2).
MPEP 2106.04(a)(2)(III)(C) “A Claim That Requires a Computer May Still Recite a Mental Process.”
Examiner is interpreting the limitations as abstract ideas implemented on a generic computer. (Step 2A Prong 1).
The claimed “transforming (i.e., determining)” or “generating” is an observation or evaluation based on the plurality of construction project documents into structured dataset, or a searchable description for an unsearchable content in the plurality of construction project documents. This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person transforms (i.e., determines) construction project documents into structured dataset or generates a searchable description for an unsearchable content in the plurality of construction project documents. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
The claimed “generating” is an observation or evaluation based on a large language model (LLM) prompt based at least in part on the question.”
This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person generates a large language model (LLM) prompt based at least in part on the question. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
The claimed “querying” and “identify “are observations or evaluations based respectively on one or more LLMs and one or more relevant documents from the plurality of construction project documents and one or more relevant chunks related to the question.
This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person query one or more LLMs and identifies using one or more LLMs, one or more relevant documents from the plurality of construction project documents and one or more relevant chunks related to the question. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
The claimed “generating “is an observation or evaluation based respectively on a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents.
This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person generates a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
The newly added claim features do not improve the functionality of a computer or any technology (see rejection above).
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of “message interface” to perform the claim steps amounts to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claim is not patent eligible.
Argument (page 19)
Applicant appears to assert that like the subject matter at issue in Enfish and Visual Memory, the subject matter of instant claim 1 provides technical improvement over the Retrieval-Augmented Generation (RAG) process in the field of querying information about construction project documents.
Particularly, the claimed invention recites "generating embeddings of a plurality of construction project documents at least by: i) transforming an unstructured data in the plurality of construction project documents into structured dataset, or generating a searchable description for an unsearchable content in the plurality of construction project documents, and ii) applying an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents" (emphasis added).
Accordingly, like Enfish and Visual Memory, the instant claims recite patent eligible
subject matter since they are directed to improvements in computer capabilities. Applicant respectfully requests that the Office withdraw the Section 101 rejection
Examiner's response
Examiner respectfully disagrees. The claimed “applying “is an observation or evaluation based respectively on an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents, wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project.”
This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person applies an embedding model to the structured dataset, the searchable description or at least a portion of the plurality of the construction project documents, wherein the plurality of construction project documents comprise at least one of a specification, a submittal, a Request for Information (RFI), a change order, a communication message related to a construction project, a blueprint or a drawing related to a construction project. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
Examiner respectfully disagrees. The claimed “generating “is an observation or evaluation based respectively on embeddings of a plurality of construction project documents. This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person generates embeddings of a plurality of construction project documents. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
Argument (page 20)
Applicant appears to assert that the claimed invention beneficially improves information retrieval accuracy by "generating a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents, and generating an answer based at least in part on the context data."
Accordingly, Applicant respectfully submit that the present claims are patent-eligible and
that the § 101 rejections of the claims be withdrawn.
Examiner's response:
Examiner respectfully disagrees. The claimed “generating “is an observation or evaluation based respectively on a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents.
This type of observation or evaluation is an act that can be practically performed in the human mind, similar to the mental thought processes that occur when a person generates a context data by recursively summarizing the identified one or more relevant chunks across the one or more documents. Such mental observations or evaluations fall within the “mental processes” grouping of abstract idea set forth in the 2019 PEG. 2019 PEG Section I, 84 Fed. Reg. at 52.
Examiner interpreted this limitation as an observation. See MPEP 2106.04(a), particularly MPEP 2106.04(a)(2)(III)(C).
For the dependent claims 2-19, no further arguments were presented, Examiner respectfully maintains the 35 USC 101 rejections upon them, due to their nature of dependence upon their respective independent claims 1 and 20.
Therefore, claims 1-20 are directed to an abstract idea and, as such, the rejection under 35 U.S.C. §101 is maintained.
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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABABACAR SECK whose telephone number is (571)270-7146. The examiner can normally be reached Monday-Friday 8:00 A.M.-6:00 P.M..
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, Lamardo Viker, can be reached on 570-270-5871. 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.
/ABABACAR SECK/Examiner, Art Unit 2147
/VIKER A LAMARDO/Supervisory Patent Examiner, Art Unit 2147
Prosecution Insights