DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
In 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 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.
Status of Claims
Claims 1-5 and 7-23 remain pending, and are rejected.
Claim 6 has been cancelled.
Response to Arguments
Applicant’s arguments filed on 5/26/2026 with respect to the rejection under 35 U.S.C. 101 have been fully considered, but are not persuasive for at least the following rationale:
Applicant’s arguments filed on 5/26/2026 with respect to the rejection under 35 U.S.C. 101 for claims directed to a judicial exception are not persuasive.
Notably, on page 21 of the Applicant’s Remarks, arguments are made that the claims have been amended such that the decoder limitation recites a specific, named computational mechanism, such as an attention operation that generates keys, queries, and values from a corresponding portion of the prompt to generate an attention output, that are a technically defined mathematical transformation that is characteristic of the GPT-style transformer decoder architecture. On pages 21-22, the Applicant argues that the prompt generation limitation recites generating a prompt structured as a sequence of input tokens, which is argued to be a specific input format, and the output satisfies the constraint for each candidate replacement item, tying the token generation process to a specific, redefined constraint, such that the output of the decoder is not merely an abstract recommendation. The claims as a whole are argued to recite a specific improvement in the functioning of a computer system itself, to automatically transform a database object (original cart) into a new database object (replacement cart) in which every component satisfies a predetermine attribute constraint through a technical interplay of the prompting module, token-based input encoding, and the attention-mechanism-based decoder architecture of the language model. On page 23, it is argues that the amended claims new recite generating keys, queries, and values to produce an attention output, defining how each decoder processes its corresponding portion of the prompt. On pages 23-24, arguments are made that the large language model is now defined as having at least 1 billion parameters, the decoders having a GPT architecture, which is a specific, named transformer variant, and the sequence of tokens arranged as a tensor, the tensor representation being a specific data structure that is integral to the transformer’s computational process.
Examiner respectfully disagrees. The present claims recite performing an attention operation that generates keys, queries, and values to generate an attention output, but does not recite any detail as to how the attention operation I performed, how the keys, queries, and values are generated, or how they are used to generate a corresponding output token. The claims merely recite that the language model is includes a set of decoders that are configured as GPT architecture without reciting any detail as to how they are configured as GPT architecture and any underlying technology of how these operations are performed. Furthermore, the specification does not disclose any of these details, having disclosure, such as in specification paragraph [0035], that does not disclose more detail than the transformer having a GPT architecture including a set of decoders to perform one or more operations to input data. The recitation of generating a prompt structured as a sequence of input tokens arranged as a tensor having a first dimension representing a number of tokens, a second dimension representing a sample in a batch of input data, and a third dimension representing a space in an embedding space are also not recited with any particularity. These limitations merely represent how information is organized, and merely adds a little more detail to the mere input of abstract data into a model to generate an output. How the tokens are generated, the specific technical process of how the model functions is not recited in the claims. The claims merely flesh out generic model operations that merely describe an input and output of data of the abstract idea, without changing or improving any ability of how any machine learning, neural network, or similar technology functions.
Satisfying the constraint for each candidate replacement item, tying the token generation process to a specific, redefined constraint does not represent any technical performance, but merely details the abstract idea, such that the replacement items are associated with a value of an attribute than is lower than a value of the attribute of the item from the original set of items. This is a part of the abstract process of determining which items to include as a candidate set of replacement items. The transformation of database objects of an original cart to a replacement cart also does not represent a change in data structure, such as how a computer retrieves and stores data in memory, but represents the abstract process of merely changing the items in the cart to be replacement items, when the original item in unavailable to the consumer. The claims are directed to the process of receiving an order with items that are unavailable, scoring and identifying items that can replace the unavailable items, predicting a likelihood of conversion, and replacing the unavailable items with the replacement items. The additional elements of the prompt, model, and input/output tokens merely receive an input of information of the abstract idea, and provide an output. The amendments merely add more detail to how the prompt is organized and flesh out generic details of the models, but do not actually recite any technical functionality of how the models are being changed or improved. The claims merely recite the generic use of various elements of what type of elements are used. Specification paragraph [0032] also discloses that the input data or output data may be configured as any number of dimensions depending on the type of data, showing that the various elements of the input tokens and tensors are not a focus of the claims, but merely an added element to provide a general link to a computing environment, and does not change how any computer element itself functions.
In view of the above, the rejection under 35 U.S.C. 101 has been maintained below.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-5 and 7-23 are rejected under 35 U.S.C. 101 because the claims are directed to a judicial exception without significantly more.
Step 1:
Claims 1-5 and 7-13 are directed to a method, which is a process. Claims 14-19 are directed to a computer program product, which is an article of manufacture. Claims 20-23 are directed to a computer system, which is an apparatus. Therefore, claims 1-5 and 7-23 are directed to one of the four statutory categories of invention.
Step 2A (Prong 1):
Taking claim 20 as representative, claim 20 sets forth the following limitations (emphasized in bold) reciting the abstract idea of identifying a replacement product for an unavailable item in an order:
accessing an order of a user of an online system, the order comprising an original set of items;
accessing a replacement model integrated into the computer system, wherein the replacement model is a machine-learning model trained to identify a set of candidate replacement items for an item from the original set of items;
applying the replacement computer model to identify, based at least in part on a replacement score for each item of a plurality of items, the set of candidate replacement items from the plurality of items;
selecting a subset of candidate replacement items from the identified set of candidate replacement items, based at least in part on a constraint that each candidate replacement item in the subset of candidate replacement items is associated with a value of an attribute that is lower than a value of the attribute of the item from the original set of items, wherein selecting the subset of candidate replacement items comprises:
generating, by a prompting module of the computer system, a prompt for input into a language model integrated into the computer system, the language model being having at least 1 billion parameters in a deep neural network, the prompt including information about a type of each candidate replacement item from the identified set of candidate replacement items, the value of the attribute of each candidate replacement item, and a size of each candidate replacement item, the prompt being structured as a sequence of input tokens encoding item-specific features for each candidate replacement item, wherein the sequence of input tokens is arranged as a tensor having a first dimension representing a number of tokens, a second dimension representing a sample number in a batch of input data, and a third dimension representing an embedding space;
requesting the language model to generate, using the prompt, a response including the subset of candidate replacement items satisfying a constraint that each candidate replacement item is associated with a value of the attribute that is lower than a value of the attribute of the item from the original set of items, the language model including a set of decoders configured as a generative pre-training (GPT) architecture, each decoder from the set of decoders performing an attention operation that generates keys, queries, and values from a corresponding portion of the prompt to generate an attention output, and performing one or more operation to the attention output to generate an output token of a sequence of output tokens, the prompt including a sequence of input tokens, and the sequence of output tokens forming the response generated by the language model;
accessing a conversion model integrated into the computer system, wherein the conversion model is a machine-learning model trained to identify, based on a predicted likelihood of conversion by the user for each candidate replacement item in the subset of candidate replacement items, a candidate replacement item from the subset of candidate replacement items;
applying the conversion model to identify, based at least in part on a conversion score for each candidate replacement item in the subset of candidate replacement items, a set of replacement items;
automatically updating a database of the online system that stores the original set of items in the order with the set of replacement items;
causing a device associated with a user of the online system to display a user interface with a cart including the set of replacement items.
The recited limitations above set forth the process for identifying a replacement product for an unavailable item in an order. These limitations amount to certain methods of organizing human activity, including commercial or legal interactions (e.g. advertising, marketing or sales activities or behaviors, etc.). The claims recite steps for finding replacement items that are similar to an original item in an order that are likely to be purchased by the user (see specification [0001-0002] disclosing the problem of suggesting alternative items and automating manual human input), which is a sales and marketing activity.
Such concepts have been identified by the courts as abstract ideas (see: 2106.04(a)(2)).
Step 2A (Prong 2):
Examiner acknowledges that representative claim 20 recites additional elements, such as:
a processor;
a non-transitory computer-readable storage medium having instructions, that when executed by the processor, cause the computer system to perform steps;
an online system;
a replacement model integrated into the computer system, wherein the replacement model is a machine-learning model;
a prompting module of the computer system;
the prompt being structured as a sequence of input tokens for each candidate replacement item, wherein the sequence of input tokens is arranged as a tensor having a first dimension representing a number of tokens, a second dimension representing a sample number in a batch of input data, and a third dimension representing an embedding space;
the language model including a set of decoders configured as a generative pre-training (GPT) architecture, each decoder from the set of decoders performing an attention operation that generates keys, queries, and values from a corresponding portion of the prompt to generate an attention output, and performing one or more operation to the attention output to generate an output token of a sequence of output tokens, the prompt including a sequence of input tokens, and the sequence of output tokens forming the response generated by the language model;
accessing a conversion model integrated into the computer system, wherein the conversion model is a machine-learning model;
a database of the online system;
causing a device of a user of the online system to display a user interface;
Taken individually and as a whole, representative claim 20 does not integrate the recited judicial exception into a practical application of the exception. The additional elements do no more than generally link the use of a judicial exception to a particular technological environment or field of use.
Secondly, this is also because the claim fails to (i) reflect an improvement in the functioning of a computer, or an improvement to other technology or technical field, (ii) implement the judicial exception with, or use the judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, (iii) effect a transformation or reduction of a particular article to a different state or thing, or (iv) applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment.
Although the claims recite a processor and a non-transitory computer-readable storage medium, these elements are recited with a very high level of generality. Specification paragraph [0100] merely discloses the processor comprising one or more processors or processing units. Specification paragraph [0101] merely discloses the non-transitory computer-readable medium as including any embodiment of a computer program product or other data combination. As such, it is evident that these are not any particular devices or otherwise integral to the claims other than to provide a general link to a computing environment, such that the abstract idea may be implemented on a computing device. The online system is described in specification paragraph [0042] and in Figure 2, which describes various modules to perform the steps of the abstract idea. It is clear that the online system is just software to perform the calculations and steps of the abstract idea, and does not represent anything more than the abstract idea being performed over a network. The machine-learning models are described in paragraph [0102], which provides a very general description of machine-learning models without any particular description as to how they function at a technical level. The user device is disclosed in paragraph [0011], which discloses that it may be any personal or mobile computing device, such as a smartphone, laptop computer, desktop computer, etc. It is clear that the additional elements are generally applied to the abstract idea, and merely provide a general link to a computing environment. The decoders are also recited with a very high level of generality, the claims merely reciting performing one or more operations to a corresponding portion of the prompt and including input and output tokens. The specification also does not disclose any further detail except reiterating the claim limitation in paragraph [0035]. The machine learning models are also not disclosed with any particularity, the specification merely disclosing the machine-learning models are language models configured to perform one or more natural language processing tasks (specification: [0030]). Paragraph [0059] also lists that the machine-learning model may include regression models, support vector machines, naïve bayes, etc. The architecture of GPT is also only recited once in the claims in specification paragraph [0034], and merely discloses that the transformer has a GPT architecture. There are not particular functionalities of the GPT architecture or any changes to how it technically functions in the claims or the specification. It is evident that these models are any generic machine-learning models that are merely applied to the abstract idea to perform calculations of the abstract idea, and only provide a general link to a computing environment.
In view of the above, under Step 2A (prong 2), claim 20 does not integrate the recited exception into a practical application (see again: MPEP 2106.04(d)).
Step 2B:
Returning to representative claim 20, taken individually or as a whole, the additional elements of claim 20 do not provide an inventive concepts (i.e. whether the additional elements amount to significantly more than the exception itself). As noted above, the additional elements recited in representative claim 20 are recited in a generic manner with a high level of generality and only serve to implement the abstract idea on a generic computing device. The claims result only in an improved abstract idea itself and do not reflect improvements to the functioning of a computer or another technology or technical field. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements used to perform the claimed process ultimately amount to no more than the mere instructions to apply the exception using a generic computer and/or no more than a general link to a technological environment.
Even when considered as an ordered combination, the additional elements of claim 20 do not add anything further than when they are considered individually.
In view of the above, representative claim 20 does not provide an inventive concept under step 2B, and is ineligible for patenting.
Regarding Claim 1 (method): Claim 1 recites at least substantially similar concepts and elements as recited in claim 20 such that similar analysis of the claims would be readily apparent to one of ordinary skill in the art. As such, claims 1 is rejected under at least similar rationale as provided above regarding claim 20.
Regarding Claim 14 (computer program product): Claim 14 recites at least substantially similar concepts and elements as recited in claim 20 such that similar analysis of the claims would be readily apparent to one of ordinary skill in the art. As such, claims 14 is rejected under at least similar rationale as provided above regarding claim 20.
Regarding Claim 21 (method): Claim 21 recites at least substantially similar concepts and elements as recited in claim 20 such that similar analysis of the claims would be readily apparent to one of ordinary skill in the art. As such, claims 21 is rejected under at least similar rationale as provided above regarding claim 20.
Dependent claims 2-5, 7-13, 15-19, and 22-23 recite further complexity to the judicial exception (abstract idea) of claim 20, such as by further defining the algorithm for identifying a replacement product for an unavailable item in an order. Thus, each of claims 2-5, 7-13, 15-19, and 22-23 are held to recite a judicial exception under Step 2A (Prong 1) for at least similar reasons as discussed above.
Under prong 2 of step 2A, the additional elements of dependent claims 2-5, 7-13, 15-19, and 22-23 also do not integrate the abstract idea into a practical application, considered both individually or as a whole. More specifically, dependent claims 2-5, 7-13, 15-19, and 22-23 rely on at least similar elements as recited in claim 20. Further additional elements are also acknowledged; however, the additional elements of claims 2-5, 7-13, 15-19, and 22-23 are recited only at a high level of generality (i.e. as generic computing hardware) such that they amount to nothing more than the mere instructions to implement or apply the abstract idea on generic computing hardware (or, merely uses a computer as a tool to perform an abstract idea). Further, the additional elements do no more than generally link the use of a judicial exception to a particular technological environment or field of use (such as the Internet or computing networks).
Secondly, this is also because the claims fails to (i) reflect an improvement in the functioning of a computer, or an improvement to other technology or technical field, (ii) implement the judicial exception with, or use the judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim, (iii) effect a transformation or reduction of a particular article to a different state or thing, or (iv) applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment.
Taken individually and as a whole, dependent claims 2-5, 7-13, 15-19, and 22-23 do not integrate the recited judicial exception into a practical application of the exception under step 2A (prong 2).
Lastly, under step 2B, claims 2-5, 7-13, 15-19, and 22-23 also fail to result in “significantly more” than the abstract idea under step 2B. The dependent claims recite additional functions that describe the abstract idea and use the computing device to implement the abstract idea, while failing to provide an improvement to the functioning of a computer, another technology, or technical field. The dependent claims fail to confer eligibility under step 2B because the claims merely apply the exception on generic computing hardware and generally link the exception to a technological environment.
Even when viewed as an ordered combination (as a whole), the additional elements of the dependent claims do not add anything further than when they are considered individually.
Taken individually or as an ordered combination, the dependent claims simply convey the abstract idea itself applied on a generic computer and are held to be ineligible under Steps 2B for at least similar rationale as discussed above regarding claim 20. Thus, dependent claims 2-5, 7-13, 15-19, and 22-23 do not add “significantly more” to the abstract idea.
Subject Matter Free of Prior Art
The claims have been determined to be free of the prior art for the reasons as indicated in the previous Office Action mailed on 3/17/2026.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TIMOTHY J KANG whose telephone number is (571)272-8069. The examiner can normally be reached Monday - Friday: 8:30am - 7:00pm EST.
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, Maria-Teresa Thein can be reached at 571-272-6764. 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.
/T.J.K./Examiner, Art Unit 3689
/VICTORIA E. FRUNZI/Primary Examiner, Art Unit 3689 7/1/2026
Prosecution Insights