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 .
Priority
Receipt is acknowledged that application claims priority to foreign application with application number IN202321087441 dated 12/20/2023. Copies of certified papers required by 37 CFR 1.55 have been received. Priority is acknowledged under 35 USC 119(e) and 37 CFR 1.78.
application.
Information Disclosure Statement
The IDS dated 12/17/2024 has been considered and placed in the application file.
Claim Objections
Claim(s) 1,5,and 9 is/are objected to because of the following informalities:
Claim 1,5,and 9, cites “a first data repository, and a second data repository;” while the dependent claims cite “first repository” or “second repository”, examiner suggests the below,
Claim 1,5,and 9, line(s) 10, should be “based architecture, wherein the LLM agent-based architecture comprises a first agent and a second agent interacting with each other and the user, a first
Claim(s) 2-4,6-8, and 10-12 depend either directly or indirectly from the objection of claim(s) 1,5, and 9, therefore they are also objected.
Appropriate correction is required.
Claim 1,5,and 9, cites “, the first agent to perform a first of type of tasks” the word “of” is improper, examiner suggests the below,
Claim(s) 1,5,and 9 is/are objected to because of the following informalities:
Claim 1,5,and 9, line(s) 11, should be “enabling , via the one or more hardware processors, the first agent to perform a first
Claim(s) 2-4,6-8, and 10-12 depend either directly or indirectly from the objection of claim(s) 1,5, and 9, therefore they are also objected.
Appropriate correction is required.
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 1-12 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 5, and 9, Further claim 1 recites A processor implemented method comprising steps of:
receiving, via an input/output interface, a research proposal
document as an input from a user, wherein the research proposal document comprises text depicting a high-level description of a research problem and a motivation behind the research problem;
inputting, via one or more hardware processors, the research proposal document as a query to a large language models (LLM) agent-
based architecture, wherein the LLM agent-based architecture comprises a first agent and a second agent interacting with each other and the user, a first data repository, and a second data repository;
enabling , via the one or more hardware processors, the first agent to perform a first of type of tasks and the second agent to perform
a second type of tasks on the query using the LLM agent-based architecture; and
obtaining, via the one or more hardware processors, a modified research proposal document with a validated motivation and a set of plausible solutions addressing the research problem based on the first
type of tasks performed by the first agent and the second type of tasks performed by the second agent, wherein the validated motivation is
iteratively updated based on a plurality of gaps identified in a plurality of prior research documents addressing the motivation behind the research problem.
Further claim 5 states A system, comprising:
a memory storing instructions;
one or more communication interfaces; and
one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware
processors are configured by the instructions to:
The limitation of “receiving…”, “inputting…”, “enabling …”, and “obtaining …” , as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, a person receiving a research proposal where it contains a description of the issue and possible solution. Further the person finds documents related to the proposal and further goes through them to present the best paragraphs. Further another person goes through those paragraphs to see if the proposal already exists/ any gaps. If they are the same the second person would see the differences between the proposal and the document to further adjust the research proposal to be something new.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application. In particular, the claim only recites additional elements that are computer components “processor” (paragraph 26), “memory” (paragraphs 27), “large language models (LLM) agent” (paragraph 33 ) recited at a high-level of generality such that it amounts no more than mere instructions to apply the exception using a generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
The claims do 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 element of using the computer components 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 claims are not patent eligible.
Claims 2, 6, and 10 additionally claim 2 recite The processor implemented method of claim 1, wherein the first type of tasks performed by the first agent comprises at least one of: (i) extracting relevant information from the research proposal document, (ii) generating a plurality of relevant questions from the relevant information, (iii) retrieving a plurality of top-K research documents from the first repository having a similarity to the research proposal document using a vector representation of the research proposal document, and(iv) obtaining a plurality of paragraph chunks of each of the plurality of top-K research documents from the second repository that are created using a parser and indexer functionality of the LLM agent-based architecture. However, this limitation does not prevent a human from performing the steps mentally as described above. Further, when the person is going thru these documents he extract relevant information from each on according to the proposal. Thus, these claims are directed towards a mental process. Similar to above, no additional limitations are provided that provide a practical application, or amount to significantly more than the abstract idea. Therefore, the claims are not patent eligible.
Claims 3, 7, and 11 additionally claim 3 recites The processor implemented method of claim 1, wherein the second type of tasks performed by the second agent comprises at least one of (i) identifying a plurality of gaps in the plurality of prior research documents addressing the motivation behind the research problem, (ii) identifying the set of plausible solutions addressing the research problem, and (iii) re-writing the research proposal document based on the plurality of gaps identified in a plurality of prior research documents and the set of plausible solutions addressing the research problem. However, these limitations encompass a person going through the documents and seeing what the document lacks to modify the proposal. Thus, these claims are directed towards a mental process. Similar to above, no additional limitations are provided that provide a practical application, or amount to significantly more than the abstract idea. Therefore, the claims are not patent eligible.
Claims 4, 8, and 12 additionally recites The processor implemented method of claim 1, wherein the set of plausible solutions addressing the research problem is identified by: decomposing the research problem defined in the research proposal document into a plurality of sub-problems; identifying a subset of sub-problems from the plurality of sub- problems such that a set of hallucinated problems are eliminated; retrieving a plurality of top-K research documents from the first repository having a similarity to each of the subset of sub-problems based on a vector representation of the subset of sub-problems; determining a set of plausible solutions for addressing each subproblem from the subset of sub-problems by extracting one or more relevant texts from each of the plurality of top-K research documents stored and retrieved from the second repository; iteratively performing step of retrieving the plurality of top-K research documents and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions. However, these limitations encompass a person splitting the problem into sections. Where those sections would be the subset of problems. Then retrieving documents and specific sections that most align to each sub problem. Further having a list of possible solutions and choosing the best one. Thus, these claims are directed towards a mental process. Similar to above, no additional limitations are provided that provide a practical application, or amount to significantly more than the abstract idea. Therefore, the claims are not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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 1-3, 5-7 9-11,are rejected under 35 U.S.C. 103 as obvious over US 20260044684 A1, (CHUN; Jong Yoon.) in view of Wu, Qingyun, et al. "Autogen: Enabling next-gen llm applications via multi-agent conversation." arXiv preprint arXiv:2308.08155 (2023) in further view of Díaz, Oscar, and Jeremías P. Contell. "Developing research questions in conversation with the literature: operationalization & tool support." Empirical Software Engineering 27.7 (2022): 174.
Claim 1, 5, and 9
Regarding Claim 1, 5, and 9, CHUN teach
1. A processor implemented method comprising steps of:
receiving, via an input/output interface, a research proposal document as an input from a user,
(paragraph 185 "Meanwhile, this initial proposal may be created and acquired as a form prepared to include predetermined items is provided to the first mover who proposed the intended use information. For example, if the first mover connects to the server 100 using their own terminal 400, this form may be provided to the first mover's terminal 400. Then, the first mover may fill in the desired information, including the intended use information, in this form.")
wherein the research proposal document comprises text depicting a high-level description of a research problem
(paragraph 82 "The first mover provides an initial proposal to a language model. The initial proposal may be a type of information or a draft proposal. Such an initial proposal may include a request or query that 'a diagnosis is needed for the cause of a certain disease, condition, or symptom', which may be more specifically referred to as intended use information. If such a language model is implemented as a generative language model, the aforementioned initial proposal may be provided to this generative language model through the RAG (Retrieval-Augmented Generation) technique, but is not limited thereto."
Paragraph 83-93 "
Meanwhile, the aforementioned intended use information may include at least one of the following items, but is not limited thereto.
[0084] Diagnostic test purpose
[0085] Diagnostic test method
[0086] Specimen type
[0087] Target analyte information to be detected or analyzed
[0088] Pathogen information
[0089] Information on contaminants
[0090] Information about a pathogenic gene or SNP (Single Nucleotide Polymorphism)
[0091] Disease name
[0092] Infectious disease name
[0093] Symptom information")
and a motivation behind the research problem;
(paragraph 94 "The intended use information will be described in more detail. Medical professionals and others want to make an accurate diagnosis of a person visiting a medical institution. This is because only then is an accurate prescription possible. Accordingly, a medical professional wants to know things like 'if there is a certain symptom, what cause or disease it is due to, what pathogen is causing it, what the target analyte is,' and so on. What is desired to be known in this way may be an example of intended use information, but is not limited thereto.")
inputting, via one or more hardware processors, the research proposal document as a query to a large language models (LLM)
(paragraph 187 "Meanwhile, as previously described in the embodiment, a language model is used. Specifically, if this language model is provided with the initial proposal in the form of a query, it operates to acquire the specification of the kit. For this purpose, as mentioned before, the language model may be implemented as a generative language model, but is not limited thereto."
paragraph 190 "Here, such a language model according to the embodiment may be ChatGPT based on GPT-3 or GPT-4. Specifically, GPT-3 is the third-generation language prediction model in the GPT-n series created by a company called OpenAI. GPT-3 consists of 175 billion parameters, making it more than twice as large as the previous version, GPT-2, introduced in May 2020. It is part of a natural language processing (NLP) system of pre-trained language.")
obtaining, via the one or more hardware processors, a modified research proposal document
(Paragraph 180-183 "Referring to FIG. 8, a step S100 of acquiring an initial proposal including intended use information for an in-vitro diagnostic kit is performed. At this time, the type of kit is not to be interpreted as being limited to in-vitro diagnostic use.
[0181] Furthermore, a step S110 of providing the acquired initial proposal to a pre-trained language model is performed.
[0182] Furthermore, a step S120 of acquiring the specification (spec) for the kit from the language model that received the initial proposal is performed. At this time, the acquired kit specification includes at least one of a technical specification to be used for the R&D of the kit, a resource specification for the R&D, a design specification for the kit, a regulatory approval specification that the kit must meet for approval, and a performance specification for the kit.
[0183] Furthermore, a step S130 of controlling the R&D proposal for the kit to be generated to include the acquired kit specification and the intended use information is performed."
paragraph 37 "A server for automatically generating a research and development proposal according to a fourth embodiment, includes: a memory storing at least one instruction; and a processor, wherein when the at least one instruction is executed by the processor, an initial proposal including intended use information for an in vitro diagnostic kit is acquired, the acquired initial proposal is provided to a pre-trained language model, a specification (spec) for the kit is acquired from the language model that received the initial proposal, wherein the acquired specification for the kit includes at least one of a technical specification to be used for the research and development of the kit, a resource specification for the research and development, a design specification for the kit, a regulatory approval specification that is needed to meet for approval, and a performance specification for the kit, and a research and development proposal for the kit is generated to include the acquired specification for the kit and the intended use information."
Paragraph 203 "In this case, as new papers or patents are published or new technology is discovered, supplementary information for this missing item may be acquired. Then, an alarm that supplementary information has been acquired may be transmitted to the first mover or the author of the research and development proposal, or the research and development proposal may be modified or generated to include this supplementary information."
The modified proposal is the added acquired it from paragraph 183)
with a validated motivation
(paragraph 184 "Here, the intended use information may include at least one of a diagnostic test purpose, a diagnostic test method, a specimen type, target analyte information to be detected or analyzed, pathogen information, information on contaminants, information about a pathogenic gene or SNP (Single Nucleotide Polymorphism), a disease name, an infectious disease name, and symptom information."
Paragraph 217 "Meanwhile, a review of the aforementioned initial proposal or the specification acquired for the kit may be controlled to be performed. Specifically, whether to perform a review of the initial proposal may be determined based on a result of evaluating the expertise of the first mover who proposed the initial proposal. If the expertise is above a threshold, a review may be unnecessary, but if it is below this threshold, a review may be necessary. Here, the expertise of the first mover may be evaluated in various ways. For example, it may be based on an evaluation of the first mover's background knowledge. More specifically, if the first mover publishes two or more papers on related technology annually, the expertise of that first mover up to that year may be evaluated as being above the threshold, or if five or more related patents are filed annually, the expertise of that first mover up to that year may also be evaluated as being above the threshold. Of course, the method of determining or evaluating whether the expertise is above or below the threshold is varied and is not limited to any one method."
Paragraph 218 " Furthermore, a review of not only this initial proposal but also the final proposal, that is, the research and development proposal generated by the language model, may also be controlled to be performed. Here, the server 100 may perform a procedure to have this proposal delivered to a person who may perform a review, so that a review of this research and development proposal is performed. Furthermore, if a review of the research and development proposal and a review of the initial proposal are performed, the fee or cost set for the review of the research and development proposal in the server 100 may be higher than the fee or cost set for the review of the initial proposal. Therefore, if such a fee or cost is set or determined in the server 100, the server 100 may be operated to provide guidance on the criteria for determining this fee or cost.")
and a set of plausible solutions addressing the research problem
(paragraph 182 -183 "Furthermore, a step S120 of acquiring the specification (spec) for the kit from the language model that received the initial proposal is performed. At this time, the acquired kit specification includes at least one of a technical specification to be used for the R&D of the kit, a resource specification for the R&D, a design specification for the kit, a regulatory approval specification that the kit must meet for approval, and a performance specification for the kit.
[0183] Furthermore, a step S130 of controlling the R&D proposal for the kit to be generated to include the acquired kit specification and the intended use information is performed."
Paragraph 111- 124 "Among these, the 'kit's spec' may include at least one of a technical specification to be used for the R&D of the kit, a resource specification for the R&D, a design specification for the kit, a regulatory approval specification that the kit must meet to obtain approval, and a performance specification of the kit.
[0112] To examine this in detail, the technical specification may include at least one of the following.
[0113] Sampling technology
[0114] Extraction technology
[0115] Reaction setup
[0116] Target signal generation technology
[0117] Signal analysis
[0118] Information display technology
[0119] Alternatively, the aforementioned technical specification may include at least one of the following.
[0120] Oligo Type (primer only or primer and probe) or Structure
[0121] Target-Signal Generation Mechanism with Amplification
[0122] Amplification Curve Analysis
[0123] Amplification Curve Viewing
[0124] Liquid Handler Operation")
CHUN do not explicitly teach all of agent- based architecture, wherein the LLM agent-based architecture comprises a first agent and a second agent interacting with each other and the user, a first data repository, and a second data repository;
enabling , via the one or more hardware processors, the first agent to perform a first of type of tasks
and the second agent to perform a second type of tasks on the query using the LLM agent-based architecture; and
based on the first type of tasks performed by the first agent
and the second type of tasks performed by the second agent,
wherein the validated motivation is iteratively updated based on a plurality of gaps
identified in a plurality of prior research documents addressing the motivation behind the research problem.
However, Wu teach
agent- based architecture, wherein the LLM agent-based architecture comprises a first agent and a second agent interacting with each other and the user, a first data repository, and a second data repository;
(page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
Page 3 section 2.1 conversable agents "… AutoGen lets a human participate in agent conversation via human backed agents, which could solicit human inputs at certain rounds of a conversation depending on the agent configuration. …"
Retrieval-augmented User Proxy is being interpreted as the first agent
Retrieval-augmented Assistant is being interpreted as the second agent )
enabling , via the one or more hardware processors, the first agent to perform a first of type of tasks
(page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 1 " The Retrieval-Augmented User Proxy retrieves document chunks based on the embedding similarity, and sends them along with the question to the Retrieval-Augmented Assistant."
Retrieval-augmented User Proxy is being interpreted as the first agent )
and the second agent to perform a second type of tasks on the query using the LLM agent-based architecture; and
(page 2 figure 7 description " Overview of Retrieval-augmented Chat which involves two agents, including a Retrieval augmented User Proxy and a Retrieval-augmented Assistant. Given a set of documents, the Retrieval-augmented User Proxy first automatically processes documents—splits, chunks, and stores them in a vector database. Then for a given user input, it retrieves relevant chunks as context and sends it to the Retrieval-augmented Assistant, which uses LLM to generate code or text to answer questions. Agents converse until they find a satisfactory answer."
page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 2"The Retrieval-Augmented Assistant employs an LLM to generate code or text as answers based on the question and context provided. If the LLM is unable to produce a satisfactory response, it is instructed to reply with “Update Context” to the Retrieval-Augmented User Proxy."
Retrieval-augmented Assistant is being interpreted as the second agent )
based on the first type of tasks performed by the first agent
(see figure 8 where it shows a modified documentation. The modified document is the answer " Erin Hamlin carried the USA flag in the opening ceremony. " being the modification of the first agents context + question to the second agents answer being a portion of the document.
page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 1 " The Retrieval-Augmented User Proxy retrieves document chunks based on the embedding similarity, and sends them along with the question to the Retrieval-Augmented Assistant."
Retrieval-augmented User Proxy is being interpreted as the first agent )
and the second type of tasks performed by the second agent,
(see figure 8 where it shows a modified documentation. The modified document is the answer " Erin Hamlin carried the USA flag in the opening ceremony. " being the modification of the first agents context + question to the second agents answer being a portion of the document.
page 2 figure 7 description " Overview of Retrieval-augmented Chat which involves two agents, including a Retrieval augmented User Proxy and a Retrieval-augmented Assistant. Given a set of documents, the Retrieval-augmented User Proxy first automatically processes documents—splits, chunks, and stores them in a vector database. Then for a given user input, it retrieves relevant chunks as context and sends it to the Retrieval-augmented Assistant, which uses LLM to generate code or text to answer questions. Agents converse until they find a satisfactory answer."
page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 2"The Retrieval-Augmented Assistant employs an LLM to generate code or text as answers based on the question and context provided. If the LLM is unable to produce a satisfactory response, it is instructed to reply with “Update Context” to the Retrieval-Augmented User Proxy."
Retrieval-augmented Assistant is being interpreted as the second agent )
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified CHUN to incorporate the teachings of Wu to provide a “agent- based architecture, wherein the LLM agent-based architecture comprises a first agent and a second agent interacting with each other and the user, a first data repository, and a second data repository; enabling , via the one or more hardware processors, the first agent to perform a first of type of tasks and the second agent to perform a second type of tasks on the query using the LLM agent-based architecture; and based on the first type of tasks performed by the first agent and the second type of tasks performed by the second agent, ” Doing so would Achieve the most completive performance on solving tasks, as recognized by Wu. (figure 4 citation).
CHUN in view of Wu do not explicitly teach all of wherein the validated motivation is iteratively updated based on a plurality of gaps
identified in a plurality of prior research documents addressing the motivation behind the research problem.
However, Díaz teach
wherein the validated motivation is iteratively updated based on a plurality of gaps
(See figure one the cycle keeps repeating until the question clearly differentiates from prior literature
page 6 and 7 section 3.5 RQ scoping " Broadly, the RQ plays the role of a hypothesis about the originality of the research. The literature is the Acid Test. During the test, researchers follow Comparative Thinking, which can be regarded as the counterpart of GT’s “constant comparison” motto. Comparative Thinking seeks complementarities or dissimilarities with related work (Mussweiler and Posten 2012; Marzano et al. 2001). Seeking complementarities refers to positioning findings to the extant theoretical stream (aka synergistic positioning) (Ridder et al. 2014). This favors knowledge accumulation by allowing researchers to elaborate a phenomenon in greater detail. By contrast, seeking dissimilarities profiles one’s work in contrast with theories that diverge in either the problem being tackled or the solution being proposed (aka antagonistic positioning) (Ridder et al. 2014). The aim is to fine-tune the hypothesis (i.e., the RQ) by challenging the researcher’s assumptions."
page 7 section 3.5 RQ scoping "… Alternatively, RQ Scoping starts with an often incomplete RQ v0 (i.e., not all CGAR constructs are yet decided). This RQ often delivers an abundant collection of related work (RW v0) whose reading helps better tune the research question RQ v1, and this starts the cycle again. The RQ (as the software requirements) is iteratively unfolded as confronted with the literature (as the stake holder counterpart). …"
Page 8 section 4 RQ scoping at work "On the problem side, the student notices that excerpts introduce different ways for helping with the Goal: “to come up with a RQ”. She abstracts away from the wording of the excerpts, and introduces three codes: synthesizing, gap spotting and schema extraction ")
identified in a plurality of prior research documents addressing the motivation behind the research problem.
(page 7 section 3.5 RQ scoping "… Alternatively, RQ Scoping starts with an often incomplete RQ v0 (i.e., not all CGAR constructs are yet decided). This RQ often delivers an abundant collection of related work (RW v0) whose reading helps better tune the research question RQ v1, and this starts the cycle again. The RQ (as the software requirements) is iteratively unfolded as confronted with the literature (as the stake holder counterpart). …"
page 6 and 7 section 3.5 RQ scoping " Broadly, the RQ plays the role of a hypothesis about the originality of the research. The literature is the Acid Test. During the test, researchers follow Comparative Thinking, which can be regarded as the counterpart of GT’s “constant comparison” motto. Comparative Thinking seeks complementarities or dissimilarities with related work (Mussweiler and Posten 2012; Marzano et al. 2001). Seeking complementarities refers to positioning findings to the extant theoretical stream (aka synergistic positioning) (Ridder et al. 2014). This favors knowledge accumulation by allowing researchers to elaborate a phenomenon in greater detail. By contrast, seeking dissimilarities profiles one’s work in contrast with theories that diverge in either the problem being tackled or the solution being proposed (aka antagonistic positioning) (Ridder et al. 2014). The aim is to fine-tune the hypothesis (i.e., the RQ) by challenging the researcher’s assumptions.")
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified CHUN in view of Wu to incorporate the teachings of Díaz to provide a “wherein the validated motivation is iteratively updated based on a plurality of gaps identified in a plurality of prior research documents addressing the motivation behind the research problem.” Doing so would Make research focus on essential aspects of the research question, as recognized by Díaz. (abstract).
Regarding Claim 5, CHUN in view of Wu in further view of Díaz, CHUN further teach
5. A system, comprising:
a memory storing instructions;
(paragraph 154 "FIG. 3 is an exemplary configuration diagram of a server for automatically generating the R&D proposal 100 according to the embodiment. Referring to FIG. 3, the R&D proposal generation server 100 includes a communication module 110, a memory 120, and a processor 130. However, the configuration diagram illustrated in FIG. 3 is merely exemplary, and the spirit of the present disclosure is not to be interpreted as being limited by the configuration diagram illustrated in FIG. 3. For example, the R&D proposal generation server 100 may include at least one component not illustrated in FIG. 3 or may not include at least one of the components illustrated in FIG. 3." )
one or more communication interfaces; and
(paragraph 154 "FIG. 3 is an exemplary configuration diagram of a server for automatically generating the R&D proposal 100 according to the embodiment. Referring to FIG. 3, the R&D proposal generation server 100 includes a communication module 110, a memory 120, and a processor 130. However, the configuration diagram illustrated in FIG. 3 is merely exemplary, and the spirit of the present disclosure is not to be interpreted as being limited by the configuration diagram illustrated in FIG. 3. For example, the R&D proposal generation server 100 may include at least one component not illustrated in FIG. 3 or may not include at least one of the components illustrated in FIG. 3." )
one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
(paragraph 154 "FIG. 3 is an exemplary configuration diagram of a server for automatically generating the R&D proposal 100 according to the embodiment. Referring to FIG. 3, the R&D proposal generation server 100 includes a communication module 110, a memory 120, and a processor 130. However, the configuration diagram illustrated in FIG. 3 is merely exemplary, and the spirit of the present disclosure is not to be interpreted as being limited by the configuration diagram illustrated in FIG. 3. For example, the R&D proposal generation server 100 may include at least one component not illustrated in FIG. 3 or may not include at least one of the components illustrated in FIG. 3." )
Claim 5 contains limitations similar to those found in claims 1 and therefore are not patent eligible for the same reasons.
Regarding Claim 9, CHUN in view of Wu in further view of Díaz, CHUN further teach One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:
(paragraph 36 "A computer program according to a third embodiment may be stored on a computer-readable recording medium, and the computer program may be programmed to perform each step included in the method described above.")
Claim 9 contains limitations similar to those found in claims 1 and therefore are not patent eligible for the same reasons.
Claim 2, 6, and 10
Regarding Claim 2, 6, and 10, CHUN in view of Wu in further view of Díaz, CHUN further teach (i) extracting relevant information from the research proposal document, (ii) generating a plurality of relevant questions from the relevant information, (iii) retrieving a plurality of top-K research documents from the first repository having a similarity to the research proposal document using a vector representation of the research proposal document, and (iv) obtaining a plurality of paragraph chunks of each of the plurality of top-K research documents from the second repository that are created using a parser and indexer functionality of the LLM agent-based architecture.
(paragraph 7 "A method for automatically generating a research and development proposal for an in vitro diagnostic kit, performed by a server for automatically generating the research and development proposal according to a first embodiment, includes: the steps of acquiring an initial proposal including intended use information for the kit; providing the acquired initial proposal to a pre-trained language model; acquiring a specification (spec) for the kit from the language model receiving the initial proposal, wherein the acquired specification for the kit includes at least one of a technical specification to be used for the research and development of the kit, a resource specification for the research and development, a design specification for the kit, a regulatory approval specification that is needed to meet for approval, and a performance specification for the kit; and controlling to generate the research and development proposal for the kit including the acquired specification and the intended use information for the kit."
Paragraph 186 "The initial proposal may be incomplete. Among these, the identity information of the first mover, whether clinical samples are possessed, and the intended use information are required information, and if at least one of these is not included, it may be difficult to create a final proposal. Accordingly, in an embodiment, if the aforementioned required item is not included in the initial proposal, the first mover may be requested to provide a description for the missing required item."
Having intended use information and identifying the proposal fields and detecting missing required information teaches extracting relevant information.)
CHUN in view of Wu in further view of Díaz, Wu further teach.
The processor implemented method of claim 1, wherein the first type of tasks performed by the first agent comprises at least one of:
(page 2 section 1 introduction "AutoGen uses a generic design of agents that can lever age LLMs, human inputs, tools, or a combination of them. The result is that developers can easily and quickly create agents with different roles (e.g., agents to write code, execute code, wire in human feedback, validate outputs, etc.) by selecting and configuring a subset of built-in capabilities. The agent’s backend can also be readily extended to allow more custom behaviors. To make these agents suitable for multi-agent conversation, every agent is made conversable they can receive, react, and respond to messages. When configured properly, an agent can hold multiple turns of conversations with other agents autonomously or solicit human inputs at certain rounds, enabling human agency and automation. The conversable agent design leverages the strong capability of the most advanced LLMs in taking feedback and making progress via chat and also allows combining capabilities of LLMs in a modular fashion. (Section 2.1)"
page 2 section 1 introduction"… How can we develop a straightforward, unified interface that can accommodate a wide range of agent conversation patterns? In practice, applications of varying complexities may need distinct sets of agents with specific capabilities, and may require different conversation patterns, such as single- or multi-turn dialogs, different human involvement modes, and static vs. dynamic conversation. …")
see claim one for rationale.
Claim 3, 7, and 11
Regarding Claim 3, 7, and 11, CHUN in view of Wu in further view of Díaz, CHUN further teach
at least one of (i) identifying a plurality of gaps in the plurality of prior research documents addressing the motivation behind the research problem, (ii) identifying the set of plausible solutions addressing the research problem, and (iii) re-writing the research proposal document based on the plurality of gaps identified in a plurality of prior research documents and the set of plausible solutions addressing the research problem.
(paragraph 182 -183 "Furthermore, a step S120 of acquiring the specification (spec) for the kit from the language model that received the initial proposal is performed. At this time, the acquired kit specification includes at least one of a technical specification to be used for the R&D of the kit, a resource specification for the R&D, a design specification for the kit, a regulatory approval specification that the kit must meet for approval, and a performance specification for the kit.
[0183] Furthermore, a step S130 of controlling the R&D proposal for the kit to be generated to include the acquired kit specification and the intended use information is performed."
Paragraph 111- 124 "Among these, the 'kit's spec' may include at least one of a technical specification to be used for the R&D of the kit, a resource specification for the R&D, a design specification for the kit, a regulatory approval specification that the kit must meet to obtain approval, and a performance specification of the kit.
[0112] To examine this in detail, the technical specification may include at least one of the following.
[0113] Sampling technology
[0114] Extraction technology
[0115] Reaction setup
[0116] Target signal generation technology
[0117] Signal analysis
[0118] Information display technology
[0119] Alternatively, the aforementioned technical specification may include at least one of the following.
[0120] Oligo Type (primer only or primer and probe) or Structure
[0121] Target-Signal Generation Mechanism with Amplification
[0122] Amplification Curve Analysis
[0123] Amplification Curve Viewing
[0124] Liquid Handler Operation")
CHUN in view of Wu in further view of Díaz, Wu further teach The processor implemented method of claim 1, wherein the second type of tasks performed by the second agent comprises
(page 2 section 1 introduction "AutoGen uses a generic design of agents that can lever age LLMs, human inputs, tools, or a combination of them. The result is that developers can easily and quickly create agents with different roles (e.g., agents to write code, execute code, wire in human feedback, validate outputs, etc.) by selecting and configuring a subset of built-in capabilities. The agent’s backend can also be readily extended to allow more custom behaviors. To make these agents suitable for multi-agent conversation, every agent is made conversable they can receive, react, and respond to messages. When configured properly, an agent can hold multiple turns of conversations with other agents autonomously or solicit human inputs at certain rounds, enabling human agency and automation. The conversable agent design leverages the strong capability of the most advanced LLMs in taking feedback and making progress via chat and also allows combining capabilities of LLMs in a modular fashion. (Section 2.1)"
page 2 section 1 introduction"… How can we develop a straightforward, unified interface that can accommodate a wide range of agent conversation patterns? In practice, applications of varying complexities may need distinct sets of agents with specific capabilities, and may require different conversation patterns, such as single- or multi-turn dialogs, different human involvement modes, and static vs. dynamic conversation. …"
page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 2"The Retrieval-Augmented Assistant employs an LLM to generate code or text as answers based on the question and context provided. If the LLM is unable to produce a satisfactory response, it is instructed to reply with “Update Context” to the Retrieval-Augmented User Proxy."
Retrieval-augmented Assistant is being interpreted as the second agent )
see claim one for rationale.
Claims 4, 8, and 12,are rejected under 35 U.S.C. 103 as obvious over US 20260044684 A1, (CHUN; Jong Yoon.) in view of Wu, Qingyun, et al. "Autogen: Enabling next-gen llm applications via multi-agent conversation." arXiv preprint arXiv:2308.08155 (2023) in view of Díaz, Oscar, and Jeremías P. Contell. "Developing research questions in conversation with the literature: operationalization & tool support." Empirical Software Engineering 27.7 (2022): 174 in view of in US 20260010963 A1, (Karmakar; Aveek.) further view of US 20210319066 A1, (Boxwell; Stephen Arthur.)
Claim 4, 8 and 12
Regarding Claim 4, 8, and 12, CHUN do not explicitly teach all of 4. The processor implemented method of claim 1, wherein the set of plausible solutions addressing the research problem is identified by:
decomposing the research problem defined in the research proposal document into a plurality of sub-problems;
identifying a subset of sub-problems from the plurality of sub-problems such that a set of hallucinated problems are eliminated;
retrieving a plurality of top-K research documents from the first repository having a similarity to each of the subset of sub-problems based on a vector representation of the subset of sub-problems;
determining a set of plausible solutions for addressing each subproblem from the subset of sub-problems
by extracting one or more relevant texts from each of the plurality of top-K research documents stored and retrieved from the second repository;
iteratively performing step of retrieving the plurality of top-K research documents
and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and
identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions.
However, Wu teach
4. The processor implemented method of claim 1, wherein the set of plausible solutions addressing the research problem is identified by:
decomposing the research problem defined in the research proposal document into a plurality of sub-problems;
(page 2 section 1 introduction "…Third, LLMs have demonstrated ability to solve complex tasks when the tasks are broken into simpler subtasks. … ")
by extracting one or more relevant texts from each of the plurality of top-K research documents stored and retrieved from the second repository;
(page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
iteratively performing step of retrieving the plurality of top-K research documents
(page 22 section detailed work flow "The workflow of Retrieval-Augmented Chat is illustrated in Figure 7. To use Retrieval-augmented Chat, one needs to initialize two agents including Retrieval-augmented User Proxy and Retrieval-augmented Assistant. Initializing the Retrieval-Augmented User Proxy necessitates specifying a path to the document collection. Subsequently, the Retrieval-Augmented User Proxy can download the documents, segment them into chunks of a specific size, compute embeddings, and store them in a vector database. Once a chat is initiated, the agents collaboratively engage in code generation or question-answering adhering to the procedures outlined below:"
page 22 section detailed work flow number 2"The Retrieval-Augmented Assistant employs an LLM to generate code or text as answers based on the question and context provided. If the LLM is unable to produce a satisfactory response, it is instructed to reply with “Update Context” to the Retrieval-Augmented User Proxy.")
See claim one for rationale.
CHUN in view of Wu, in view of Díaz do not explicitly teach all of identifying a subset of sub-problems from the plurality of sub-problems such that a set of hallucinated problems are eliminated;
retrieving a plurality of top-K research documents from the first repository having a similarity to each of the subset of sub-problems based on a vector representation of the subset of sub-problems;
determining a set of plausible solutions for addressing each subproblem from the subset of sub-problems
and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and
identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions.
However, Karmakar teach
identifying a subset of sub-problems from the plurality of sub-problems such that a set of hallucinated problems are eliminated;
(paragraph 36 provisional paragraph 34 "As noted above, in at least some embodiments, the described techniques include using examples of query-response pairs for LLM prompt generation (e.g., ReACT, or Reasoning and ACTing, query-response pair examples that each include one or more series of a reasoning activity, followed by an acting activity that is based on the results of the reasoning activity, followed by an observation activity that is based on the results of the acting activity). The ReACT LLM prompting may use LLM reasoning to break a user query into solvable sub problems using the defined tools discussed above, with the ReACT processing solving problems in steps by deciding a next action to take, based on the current observation from the tool. The generated enhanced query prompt for the LLM may include providing instructions for performing the ReACT processing, including via the provided example query-response pairs (e.g., 3 selected query-response pairs that are associated with a determined topic for the user query or that are otherwise matched to the user query, such as based on similarity between the user query and the query portion of the selected query-response pairs, or based on such similarity a combination of the user query and chatbot history for the current interaction session). The generated enhanced query prompt for the LLM may further include additional information, such as the following: formatting instructions about how to format response data (e.g., using bullet points, sections, list items, etc.); citation instructions related to citing the source of the information used in generating the response every time information from a document or tool is used (e.g., for the mortgage calculator tool, using a list of Json with the following structure ‘{“source”: “https://<web-site>/mortgage-calculator”, “content”: “ . . . ”}’); etc."
Paragraph 55 provisional paragraph 62" If it is instead determined in block 520 that there is not a match between the user query and the list of noncompliant deny phrases, or after block 523, the routine continues to block 530 to submit the query to the trained classifier model to determine whether to classify the user query as rejected or accepted. In block 535, if the classifier model determines to reject the user query, the routine continues to block 521, and otherwise continues to block 540 to modify the user query to include the predefined LLM prompt instructions to refuse to provide responses to inputs with references to defined legally protected classes. After block 540, the routine continues to block 590 to provide an indication that the user query is not rejected.")
retrieving a plurality of top-K research documents from the first repository having a similarity to each of the subset of sub-problems based on a vector representation of the subset of sub-problems;
(paragraph 22 provisional paragraph 21 "FIG. 1C continues the examples of FIGS. 1A-1B, and illustrates one example embodiment of the AQRIG LLM Prompt Generator component 148 discussed in FIG. 1A. In particular, in the illustrated embodiment, the component 148 performs various activities to generate an enhanced query prompt 158 to provide to the AQRIG LLM component 150. In operation, the component 148 receives a user query with modifications 197 from the AQRIG Fair Housing Filter component 144, and determines 162 one of multiple defined housing-related topics to which the query corresponds. In block 164, the component then compares the user query to prior queries that are mapped to documents used in prior responses to those prior queries. If it is determined in block 166 that there are one or more matching prior queries (e.g., with a similarity above a defined threshold), the routine continues to block 168 to retrieve the one or more documents mapped to those matching one or more prior queries, and otherwise continues to block 172 to check if the determined topic for the user query corresponds to one of multiple defined tools. If so, the routine continues to block 174 to retrieve information from the defined tool about that housing-related topic, and otherwise continues to block 176 to determine one or more of a defined group of documents whose contents match the query contents (e.g., with a similarity above a defined threshold), with those one or more best matching documents retrieved in block 178—in particular, in this example, the routine in block 176 encodes the contents of the query (e.g., generates a vector embedding representation of the query contents) and uses a distance metric to determine that the similarity of the encoded query contents to one or more of the encoded representations of the retrieved documents is above a defined threshold (e.g., below a defined distance) or otherwise provides a best match. After blocks 168, 174, or 178, the routine in block 180 selects one or more example query-response pairs (e.g., as matching the user query, based on the determined topic for the query, etc.). In block 182, the routine then combines the user query with the modifications 197, user data 151, selected query-response pairs, retrieved documents or information from blocks 168 or 174 or 178, and optionally one or more additional elements to generate enhanced query prompt 158—such optional additional elements may include, for example, information from an intermediate LLM query response (if any), instructions to restrict the response to the defined topic, response formatting instructions, etc. After the enhanced query prompt 158 is generated, it is provided to the AQRIG LLM component 150.")
determining a set of plausible solutions for addressing each subproblem from the subset of sub-problems
(paragraph 34 provisional paragraph 33 "As noted above, in at least some embodiments, the described techniques include using a defined group of authoritative source tools to each provide current housing-related information of a particular type used in responses (e.g., information about current housing statistics and/or individual available houses, information about current mortgage rates and/or other housing affordability factors, etc.)"
Paragraph 36 provisional paragraph 34 "As noted above, in at least some embodiments, the described techniques include using examples of query-response pairs for LLM prompt generation (e.g., ReACT, or Reasoning and ACTing, query-response pair examples that each include one or more series of a reasoning activity, followed by an acting activity that is based on the results of the reasoning activity, followed by an observation activity that is based on the results of the acting activity). The ReACT LLM prompting may use LLM reasoning to break a user query into solvable sub problems using the defined tools discussed above, with the ReACT processing solving problems in steps by deciding a next action to take, based on the current observation from the tool. The generated enhanced query prompt for the LLM may include providing instructions for performing the ReACT processing, including via the provided example query-response pairs (e.g., 3 selected query-response pairs that are associated with a determined topic for the user query or that are otherwise matched to the user query, such as based on similarity between the user query and the query portion of the selected query-response pairs, or based on such similarity a combination of the user query and chatbot history for the current interaction session). The generated enhanced query prompt for the LLM may further include additional information, such as the following: formatting instructions about how to format response data (e.g., using bullet points, sections, list items, etc.); citation instructions related to citing the source of the information used in generating the response every time information from a document or tool is used (e.g., for the mortgage calculator tool, using a list of Json with the following structure ‘{“source”: “https://<web-site>/mortgage-calculator”, “content”: “ . . . ”}’); etc.")
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified CHUN in view of Wu, in view of Díaz to incorporate the teachings of Karmakar to provide a “identifying a subset of sub-problems from the plurality of sub-problems such that a set of hallucinated problems are eliminated; retrieving a plurality of top-K research documents from the first repository having a similarity to each of the subset of sub-problems based on a vector representation of the subset of sub-problems; determining a set of plausible solutions for addressing each subproblem from the subset of sub-problems ” Doing so would Provide improve the identification and use responsive information to specified queries, as recognized by Karmakar. (Paragraph 39).
CHUN in view of Wu, in view of Díaz, in further view of Karmakar do not explicitly teach all of and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and
identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions.
However, Boxwell teach
and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and
(paragraph 16 "In order to address the issue of dealing with differently phrased reference documents rich with pragmatic implications in order to answer such input questions, the illustrative embodiments provide mechanisms for domain adaptation techniques that may be used to map broad references within a corpus of reference documents to a set of specific questions, where the correct answers to the specific questions are assumed to be correct answers for to the input question. That is, the illustrative embodiment employs a set of sub-questions that are identified as being related to the input question. The QA system then operates to answer each sub-question separately and merge the results, i.e. the answer to each sub-question into a single report so the most relevant information is easily accessible. By performing such operations, the illustrative embodiments improve the answering capabilities of QA systems such that differently phrased reference documents rich with pragmatic implications from the corpus of reference documents provide an accurate answer to the input question."
Paragraph 60 "Utilizing this parsing information, sub-question identification engine 124 identifies a list of sub-questions from corpus of sub-questions 126 related to input question 122 based on the parsing information, thereby forming a set of questions to be posed to the corpus of data/information 140 in order to generate one or more hypotheses. The set of questions are generated in any known or later developed query language, such as the Structure Query Language (SQL), or the like. The set of questions are applied to one or more databases storing information about the electronic texts, documents, articles, websites, and the like, that make up the corpora of data/information 140. That is, these various sources themselves, different collections of sources, and the like, represent a different corpus 142 within the corpora 140. There may be different corpora 142 defined for different collections of documents based on various criteria depending upon the particular implementation. For example, different corpora may be established for different topics, subject matter categories, sources of information, or the like." )
identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions.
(paragraph 90 ""In another operation, as result processing engine 393 adds each answer to answer data structure 394, result processing engine 393 updates each passage index for each answer occurrence of the answer to m[ai][pi], where ai is the index of the answer in answer data structure 394 and pi is the original passage index for this answer in merged passage index 396. If result processing engine 393 determines that an answer to a question from the set of questions already exists in answer data structure 394, result processing engine 393 determines whether an assigned score associated with that answer being added is higher than the answer already existing in answer data structure 394. The score associated with each answer is provided by the hypothesis and evidence scoring stage 350 and the synthesis stage 360 of QA pipeline 300. If result processing engine 393 determines that a score associated with the answer being added is higher, then result processing engine 393 sets the existing answer's score in answer data structure 394 to the higher score. If result processing engine 393 determines that the answer's score being added fails to be higher, then result processing engine 393 leaves the score of the existing answer as is. If result processing engine 393 determines that an answer to a question from the set of questions fails to exist in answer data structure 394, result processing engine 393 adds the answer with its score to answer data structure 394. Once all answers are added to answer data structure 394, result processing engine 393 sorts answer data structure 394 by score. Result processing engine 393 then generates a factoid question response report with answers from answer data structure 394 and passages from passage data structure 395. Result processing engine 393 then provides the factoid question response report to the confidence merging and ranking stage 370 so as to replace the normal operations provided by the confidence merging and ranking stage 370 and thus, be utilized in a final set of candidate answers and confidence scores stage 380 to produce a final answer and confidence score, or final set of candidate answers and confidence scores, which are output to the submitter of the original input question via a graphical user interface or other mechanism for outputting information.)
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified CHUN in view of Wu, in view of Díaz, in further view of Karmakar to incorporate the teachings of Boxwell to provide a “and identifying the set of plausible solutions to generate a consolidated list of a subset of similar sub-problems and corresponding set of plausible solutions; and identifying the set of plausible solutions addressing the research problem using the consolidated list of the subset of similar sub-problems and the corresponding set of plausible solutions.” Doing so would Solve problems with high accuracy and resilience on a large scale, as recognized by Boxwell. (Paragraph 28).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALI M HASSAN whose telephone number is (571)272-5331. The examiner can normally be reached Monday - Friday 8:00am - 4:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Paras Shah can be reached at (571)270-1650. 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.
/ALI M HASSAN/ Examiner, Art Unit 2653
/Paras D Shah/ Supervisory Patent Examiner, Art Unit 2653
07/22/2026
Prosecution Insights