DETAILED ACTION
Introduction
1. This office action is in response to Applicant's submission filed on 05/30/2024. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Claims 1-20 are currently pending and examined below.
Drawings
2. The drawings filed on 05/30/2024 have been accepted and considered by the Examiner.
Information Disclosure Statement
3. The Information Statement (IDS) filed on 05/30/2024 has been accepted/considered and is in compliance with the provisions of 37 CFR 1.97.
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.
4. Claims 1-20 are rejected under 35 U.S.C. 101 as being nothing more than an abstract idea. Regarding claim 1, the limitations of receiving a query from a user, evaluating the query to generate query evaluation results, generating an action plan for the query based on the query evaluation results, generating environment information, generating knowledge information based on well-established facts and business logic, generating memory information based on short term information and inference information; determining logical inferences about the query based on the query evaluation results, the environment information, the knowledge information, and the memory information, generating a response to the query, determining whether the response answers the query, modifying the response and generate a final response based on determining that the response answers the query and providing the final response to the user device: all fall under the category of mental processes. These steps are drafted at a high level of generality without tying them to a specific technological improvement. More specifically, these steps could be performed by a human being in his mind with the aid of (at most) a pen and a paper while listening to a user orally make a query, but for the recitation of generic computer components, and thus they fall within the --Mental Processes-- grouping of abstract ideas. Accordingly, this claim recites an abstract idea.
This judicial exception is not integrated into a practical application because the
recitation of a device, a system, processor and/or a computer readable medium merely read to generalized computer components, based upon the claim interpretation wherein the structure is interpreted using the specification. These recitations read to a set of rules learned by a human for how to perform the specific tasks. Accordingly, these additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea. 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 the integration of the abstract idea into a practical application, the additional element of using generalized computer components to generate, extract, determine, and generate, 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 therefore not patent eligible.
Claims 2-7, only provide certain details of the mental processes outlined above, such as storing the final result, reusing the final result or withholding the final result etc. These are all steps which too can be performed by a human being with (at most) the aid of a pen and paper and hence also do not amount to significantly more than the judicial exception. Claims 8-14, are device claims corresponding to method claims 1-7 and hence are also rejected at least for the reasons outlined above. Claims 15-20, are computer readable medium (CRM) claims corresponding to method claims 1-7 and hence are also rejected at least for the reasons outlined above.
Claim Rejections - 35 USC § 102
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 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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
5. Claims 1-20 are rejected under 35 U.S.C. 102 (a) (2) as being anticipated by Zang (U.S. Patent Application Publication # 2025/0147991 A1).
With regards to claim 1, Zang teaches a method, comprising receiving, by a device, a query from a user device (See figure 2, block 202 and figure 3, block 302);
evaluating, by the device, the query to generate query evaluation results (Para 27, teaches that the user query or questions can be in one form and might need to be converted to another form, e.g. by being input as text or converted from speech, e.g., via automatic speech recognition engines);
generating, by the device, an action plan for the query based on the query evaluation results (Para 28, teaches that the user query or questions will be subject to preprocessing, such as stripping out HTML, tokenization of the text, removing punctuation, lowercasing all tokens, stemming, embedding, processing all tokens based on primary search language, adding synonyms for matching tokens, etc.);
utilizing, by the device, a tools module to generate environment information based on application programming interfaces, function calls, and terminal access (Para 90, teaches that the cloud computing environments can interface with the virtualized network function cloud via APIs that expose functional capabilities of the VNEs to provide the flexible and expanded capabilities to the virtualized network function cloud. In particular, network workloads may have applications distributed across the virtualized network function cloud and cloud computing environment and in the commercial cloud or might simply orchestrate workloads supported entirely in NFV infrastructure from these third-party locations);
utilizing, by the device, a knowledge module to generate knowledge information based on well-established facts and business logic (Paragraphs 18-26, teach a knowledge base that may take a form of a self-serving online library of information about a product, service, a subject, etc. Many other functionalities of the knowledge base are also outlined);
utilizing, by the device, a memory module to generate memory information based on short term information and inference information (Para 26, teaches search and content application program interface, a vector database, etc., can be used to retrieve a list of relevant answers with the retriever. Other search methods, tools, engines, etc., can be used to retrieve a list of relevant answers with the retriever);
utilizing, by the device, an intuition module to determine logical inferences about the query based on the query evaluation results, the environment information, the knowledge information, and the memory information (Para 30, teaches a vector space model representing text documents as vectors of identifiers and is used in information retrieval, indexing and relevancy rankings. Probabilistic modeling uses a statistical technique used to consider the impact of random events or actions in predicting the potential occurrence of future outcomes such as weather forecasting and postal delivery. Neural information retrieval methods use neural networks to rank search results in response to a query. Dense passive retrieval methods use a dense passage retriever for fetching relevant passages with regard to questions asked based on the similarity between high-quality low-dimensional continuous representation of passages and questions);
processing, by the device, the action plan for the query and the logical inferences about the query, with a large language model, to generate a response to the query (Para 29, teaches that as a result of searching the relevant content from the knowledge base, a list of top contents or documents relevant to the user query is retrieved by the retriever. Para 34, teaches a first re-ranker that receives the list of retrieved documents as one input and the output of the large language model reader i.e., the generated answer as another input);
determining, by the device, whether the response answers the query (Para 43, teaches a second re-ranker that performs re-ranking of each candidate document in the list of documents against the user query or question. The list of documents includes the K number of candidate documents and at the second re-ranker, the K number of candidate documents is reranked based on similarity with respect to the user query or question. In other words, the second re-ranker does not change or reduce the first number of documents. A second result filter is arranged past the second re-ranker and configured to reduce the first number of documents to a third number of documents. The second result filter is configured to set a threshold similarity score such that candidate documents having the similarity scores lower or higher than the threshold similarity score can be filtered to be the third number of documents. Operators or users of the system are enabled to select and configure the predetermined number or the threshold similarity score in light of various factors such as business needs, customer basis and needs, etc.);
utilizing, by the device, a reflect module to modify the response and generate a final response based on determining that the response answers the query (Para 46, teaches that as a result of the grading, the first and the second reduced sets of documents are presented, in the order or scale of grading, as citations or document responses in the final response);
and providing, by the device, the final response to the user device (Para 46, further teaches that the final response is generated which includes the generated answer from the LLM reader as a generative response and the citations or document responses from the LLM grader in support of the generated answer).
With regards to claim 2, Zang teaches the method of claim 1, further comprising generating one or more tasks for the tools module to perform based on determining that the response fails to answer the query (Para 71, teaches that the LLM grader is applied by using a prompt or criteria that grade the first and the second reduced number of documents. The third answer and the fourth answer are graded based on prompts or criteria. For instance, the prompts can be such as “Your task is to compare the candidates by checking which one answer the input question [query] better. Grade each of candidates using scale 1 to 10, with 1 as the worst and 10 as the best. Output the answer and candidate documents with grades.” Using the prompt or criteria further includes setting the prompt or criteria that grade the first and the second reduced number of documents against the user input).
With regards to claim 3, Zang teaches the method of claim 1, further comprising, based on determining that the response fails to answer the query preventing utilization of the reflect module to modify the response (Para 68, teaches that the first answer is provided, as an additional route, to the re-ranking of the first answer so as to forgo to use, in the additional route, the large language model reader);
and preventing provision of the final response to the user device (Para 68, further teaches that as the analysis of the context by the LLM reader is not performed, the content of Document 3 may be disregarded or not considered as being more relevant than Document 2).
With regards to claim 4, Zang teaches the method of claim 1, wherein the action plan for the query includes a plan to solve a problem posed by the query (Para 48, teaches an example plan of solving the query regarding list of organisms or animals with the largest lifespan).
With regards to claim 5, Zang teaches the method of claim 1, further comprising storing the final response in a data structure for use as context for future queries received from the user device (Para 18, teaches that the knowledge base may include a collection of interlinked information and knowledge in a way that enables storage, analysis and reuse of the knowledge in a machine accessible way. Reuse is only possible if final responses are also stored).
With regards to claim 6, Zang teaches the method of claim 1, further comprising utilizing the tools module, the knowledge module, the memory module, and the intuition module with one or more other large language models that are different than the large language model (Para 17 and figure 1, teach the use of not one but two LLMs, an LLM reader and an LLM grader).
With regards to claim 7, Zang teaches the method of claim 1, further comprising training the large language model with the memory information (Para 38, teaches that the LLMs can be trained on a massive amount of data and enabled to perform various tasks such as answering questions in natural language. Para 39, teaches that prompts are instructions to the LLM reader. Prompts can be tuned to train the LLM reader to perform specific tasks).
With regards to claims 8-14, these are system claims for the corresponding method claims 1-7. These two sets of claims are related as method and apparatus of using the same, with each claimed system element's function corresponding to the claimed method step. Accordingly, claims 8-14 are similarly rejected under the same rationale as applied above with respect to method claims 1-7.
With regards to claims 15-20, these are CRM claims for the corresponding method claims 1-7. These two sets of claims are related as CRM and method of using the same, with each claimed method function corresponding to the claimed CRM step. Accordingly, claims 15-20 are similarly rejected under the same rationale as applied above with respect to method claims 1-7.
Conclusion
6. The following prior art, made of record but not relied upon, is considered pertinent to applicant's disclosure: Nouri (U.S. Patent Application Publication # 2024/0038226 A1) relates to executing a task using a machine learning model based on prompt generation and collaborative interactions with a user, Hanes (U.S. Patent Application Publication # 2025/0036674 A1) pertains to context injection for improved AI response. These references are also included in the PTO-892 form attached with this office action.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. If you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). In case you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NEERAJ SHARMA whose contact information is given below. The examiner can normally be reached on Monday to Friday 8 am to 5 pm. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Pierre Louis-Desir can be reached on 571-272-7799 (Direct Phone). The fax number for the organization where this application or proceeding is assigned is 571-273-8300.
/NEERAJ SHARMA/
Primary Examiner, Art Unit 2659
571-270-5487 (Direct Phone)
571-270-6487 (Direct Fax)
neeraj.sharma@uspto.gov (Direct Email)
Prosecution Insights