DETAILED ACTION
This non-final rejection is responsive to the Request for Continued Examination filed June 10, 2026. Claims 1, 7, 10, 11, 17, and 20 are currently amended. Claims 5-6 and 15-16 are canceled. Claims 1-4, 7-14, and 17-20 are pending in this application.
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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on June 10, 2026 has been entered.
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 1-4, 7-14, and 17-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1 and 11 recite running a GA (genetic algorithm) optimization procedure that includes a plurality of iterations to identify a best combination set of reranking models, the GA optimization procedure comprising, based on the expected source documents: in each iteration: running a fitness function over each reranking model of the set of reranking models using the expected source document and a respective weight; selecting top reranking models for the best combination set from the set of reranking models based on highest fitness scores calculated by the fitness function; performing a crossover procedure and a mutation procedure to generate a next population; normalizing the weights after the crossover procedure and the mutation procedure; applying each reranking model in the best combination set to reorder the source documents; optimizing respective weights for the reranking models in the best combination set to generate a weights vector; generating, by a reciprocal rank fusion (RRF), a RRF score based on the weights vector until the RRF score meets a predetermined threshold; generating, by the RRF, a fused ranking of the source documents based on the weights vector; deploying the best combination set of reranking models and the weights vector to reorder source documents retrieved in response to subsequent user queries; and ranking the source documents according to the fused ranking to generate a reordered set of source documents.
The broadest reasonable interpretation of these steps is that the steps fall within the mathematical concepts and mental process groupings of abstract ideas because they cover math concepts and/or concepts that can be performed in the human mind, including observation, evaluation, judgment, and opinion. See MPEP 2106.04(a)(2), subsection III. For example, running a genetic algorithm procedure comprising running a fitness function over each reranking model, performing crossover and mutation, normalizing weights, optimizing a weight of each reranking model to generate a weights vector, and generating a RRD score based on the weights vector are math concepts that may also be performed mentally or manually with the aid of pen and paper. Further, selecting top reranking models; applying each reranking model to reorder source documents; generating a fused ranking of the source; deploying the best combination set of reranking models and the weights vector to reorder source documents; and ranking the source documents according to the fused ranking to generate a reordered set of source documents are steps that can be performed mentally using evaluation and observation. As such, claims 1 and 11 are directed to an abstract idea.
This judicial exception is not integrated into a practical application. The additional elements “receiving a set of reranking models, and a dataset that comprises a plurality of pairs, each pair comprising question posed by a user, and respective expected source documents responsive to the question”, “in a last iteration, returning a set of selected reranking models as the best combination set” and “returning one or more highest-ranked documents for use in generating a response to the user query by the RAG pipeline from the reordered set of source documents based on the best combination set of the reranking models, the weights vector, and the fused ranking” are mere data gathering and output recited at a high level of generality, and thus are insignificant extra-solution activity. Further, the additional elements of “applying a retriever of a retrieval-augmented generation (RAG) pipeline to the questions in the dataset to retrieve source documents based on the questions in the dataset”, “optimizing by a machine learning weight optimizer”, “deploying…in the RAG pipeline” and “ranking, by the RAG pipeline” amount to no more than mere instructions to apply the exception using a generic computer (See MPEP 2106.05(f)). Lastly, the limitation “a non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations” amounts to no more than mere instructions to implement the exception using a generic computer. Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application, and the claim is directed to the judicial exception.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above, the recitations of “receiving” and “returning” are recited at a high level of generality. These elements amount to receiving or transmitting data over a network and are well-understood, routine, conventional activity. See MPEP 2106.05(d), subsection II. As discussed above, the recitations of “applying a retriever”, “optimizing by a machine learning weight optimizer”, “deploying in the RAG pipeline” and “ranking by the RAG pipeline” amount to no more than mere instructions to apply the exception using generic computer components. Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept.
Claims 2 and 12 recite wherein the documents comprise a set of source documents responsive to the questions posed by the user. This limitation further describes the documents that are ranked and returned. As discussed above, ranking documents is a mental process.
This judicial exception is not integrated into a practical application because ranking in a RAG pipeline amounts to mere instructions to apply the judicial exception on a generic computer and also merely indicates a field of use or technological environment in which the judicial exception is performed. Returning documents represents insignificant extra-solution activity.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Returning documents is recited at a high level of generality. This element amounts to receiving or transmitting data over a network and are well-understood, routine, conventional activity. See MPEP 2106.05(d), subsection II. The recitation of “using, in a RAG pipeline…to rank documents” amounts to no more than mere instructions to apply the exception using generic computer components. Further, the limitation “using, in a RAG pipeline…to rank documents” represents a field of use. Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept.
Claims 3 and 13 recite “wherein respective ranks of the documents are different from an earlier ranking of those documents that was applied to the documents when the documents were initially received”. This limitation further describes the ranks of documents, and is thus descriptive in nature. Ranking documents is a mental process as a user can rank documents in the human mind. This judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because there are no additional elements in these claims.
Claims 4 and 14 recite “wherein each of the reranking models generates a respective sub-group of the documents that are responsive to the user query”. This limitation represents an additional element. However, this judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because using reranking models in a RAG pipeline to generate sub-groups of documents amounts to mere instructions to apply the judicial exception on a generic computer and also merely indicates a field of use or technological environment in which the judicial exception is performed. Even when considered in combination, this additional element represents mere instructions to implement an abstract idea or other exception on a computer, which does not provide an inventive concept.
Claims 7 and 17 recite wherein the best respective weights are obtained using a weight optimizing process, the weight optimizing process comprises a model configured to generate a weight vector comprising a respective weight for each ranking model in the best combination set. These limitations represent a mental process and math concepts. A user can mentally generate a weight vector comprising a weight for each reranking model. This judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional element of “performed by the machine learning weight optimizer” amounts to mere instructions to apply the judicial exception on a generic computer. Even when considered in combination, this additional element represents mere instructions to implement an abstract idea or other exception on a computer, which does not provide an inventive concept.
Claims 8 and 18 recite “wherein the weight optimizing process generates respective weights for each of the reranking models, and the best respective weights generated by the weight optimizing process are more optimal, relative to the weights generated by the GA optimization procedure”. These limitations represent a mental process and math concepts. A user can mentally obtain the best weights using a weight optimizing process that generates weights that are optimal relative to weights generated by a GA procedure. This judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because there are no additional elements in these claims.
Claims 9 and 19 recite “wherein inputs to the weight optimizing process comprise weights generated by the GA optimization procedure, and the dataset.” This limitation is a mental process. A user can mentally, or with the aid of pen and paper, use a weight optimizing process having inputs generated by GA optimization procedure and the dataset. This judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because there are no additional elements in these claims.
Claims 10 and 20 recite “wherein the weight optimizing process runs until an optimal RRF score, which meets the predetermined threshold, is obtained, and wherein the optimal RRF score is generated using the weight vector generated by the model.” This limitation represents a mental process and math concept. A user can mentally, with the aid of pen and paper, perform a weight optimizing process until an optimal RFF score is obtained and generate the optimal RRF score using the weight vector generated by the model. This judicial exception is not integrated into a practical application and the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because there are no additional elements in these claims.
Response to Arguments
Applicant's arguments filed June 10, 2026 have been fully considered but they are not persuasive. Applicant argues that the claims are directed to a specific improvement in the operation of a computer-implemented RAG system. The examiner disagrees. The claims do not appear to recite any improvement to the functioning of a computer or improvement to a technical problem. Applicant’s alleged improvement appears to be directed towards an improvement to an abstract idea that is implemented on a computer.
Applicant argues that the amended claims integrate claim limitations into a practical application by deploying the selected reranking model combination and weight vector within a RAG pipeline to reorder retrieved source documents and provide highest-ranked documents for use in generating query responses. The examiner disagrees. The step of deploying the selected reranking model combination and weight vector is a mental step. A user can mentally (with aid of pen and paper) use the selected reranking model combination and weight vector to reorder retrieved source documents and provide highest-ranked documents. Deploying the combination in a RAG pipeline is merely applying the abstract idea using a computer.
Applicant argues that the claims are directed to a specific technological process for improving document retrieval and ranking within a computer-based information retrieval architecture, and further that the claimed process improves the manner in which source documents are selected and ordered for downstream response generation within the RAG system. The examiner disagrees. The claims appear to be directed to applying mental steps and calculations using a computer, and not actually improving a specific technological process. Any alleged improvement appears to be in steps that are mentally performed. The claims do not recite any additional elements that integrate the abstract idea into a practical application or amount to significantly more than the abstract idea.
Lastly, Applicant argues that the claims use the results of any math calculations to control operation of a retrieval architecture, generate a fused ranking, reorder source documents with the RAG pipeline, and provide highest-ranked documents. However, except for the recitation of implementing in the RAG pipeline, these steps of generating, reordering and providing are mental steps and thus also directed to a judicial exception. Implementation by or in a RAG pipeline amounts to mere instructions to apply the abstract idea using a computer.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALICIA M WILLOUGHBY whose telephone number is (571)272-5599. The examiner can normally be reached 9-5:30, EST, M-F.
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, Ajay Bhatia can be reached at 571-272-3906. 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.
/ALICIA M WILLOUGHBY/Primary Examiner, Art Unit 2156 July 25, 2026
Prosecution Insights