DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is in response to the application filed on 09/12/2024.
Examiner’s Note
Examiner cites paragraphs, figures, and line numbers in the references as applied to the claims
below for the convenience of the applicant. Although the specified citations are representative of the
teachings in the art and are applied to the specific limitations within the individual claim, other passages
and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant
fully consider the references in their entirety as potentially teaching all or part of the claimed invention,
as well as the context of the passage as taught by the prior art or disclosed by the examiner. As a
disclaimer, the use of underlining in direct quotes is done by the examiner for emphasis. Direct quotes
are not originally underlined in the published references cited.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 09/12/2024 is in compliance with the provisions of 37 CFR 1.97 and is being considered by the examiner.
Claim Objections
Claims 1-20 are objected to because of the following informalities:
In line 7 of Claim 1; line 12 of Claim 8; line 9 of Claim 15, “the fine-tuned generative mode” should be corrected as --the fine-tuned generative model--.
Line 1 of claim 7 and 14, after “wherein”, --the-- should be inserted.
Claims 2-7 inherit the issue from Claim 1.
Claims 9-14 inherit the issue from Claim 8.
Claims 16-20 inherit the issue from Claim 15.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis:
Claims 1-7 are directed to a method and fall within the statutory category of processes; Claims 8-14 are directed to an apparatus and falls within the statutory category of machines; Claims 15-20 are directed to a non-transitory computer-readable medium and fall within the statutory category of articles of manufacture. Therefore, "Are the claims to a process, machine, manufacture or composition of matter?" Yes. To evaluate the Step 2A inquiry "Is the claim directed to a law of nature, a natural phenomenon or an abstract idea?" we must determine, at Step 2A Prong 1, whether the claim recites a law of nature, a natural phenomenon, or an abstract idea (see MPEP § 2106.04).
Regarding Claims 1, 8, and 15:
Step 2A Prong 1 Analysis:
The claim limitation recites, generating, [], a group of code traces, each code trace of the group of code traces corresponding to a respective algorithm, of a group of algorithms, and a corresponding input; This limitation covers performance in the mind in the form of evaluation and judgement with the assistance of pen and paper. For example, a developer, with pen and paper, may create a step-by-step description of how a solution has resulted from each algorithm in the group of algorithms. More specifically, for example a developer may solve by hand a sorting algorithm with an incremental loop, choosing an input for a variable to begin the loop at the 5th index in an array. Therefore, this limitation recites a mental process. (See MPEP § 2106.04(a)(2), subsection III).
and generating, [], one or more computer programming code statements corresponding to the computer programming code or simulate an expected output of the computer programming code; This limitation covers performance in the mind in the form of evaluation and judgement with the assistance of pen and paper. For example, a developer, with pen and paper, may create the program code needed to do the step-by-step description of how a solution has resulted from an algorithm. Or a developer can simulate the expected output of programming code; such as in the case with a sorting algorithm on an unsorted list, by going through each iteration by hand, until the expected outcome of a sorted list is completed. Therefore, this limitation recites a mental process. (See MPEP § 2106.04(a)(2), subsection III).
Step 2A Prong 2 Analysis:
via a virtual machine; This limitation recites additional elements that merely recite instructions to implement an abstract idea on a generic computer. (See MPEP § 2106.05(f)).
fine-tuning a generative model in accordance with the group of code traces; This claim recites the additional element “fine-tuning” which is merely an insignificant extra-solution activity as a post-solution to building generative models, and does not integrate the judicial exception into a practical application (see MPEP § 2106.05(g)), analyzed further below in Step 2B as being well-understood, routine, and conventional.
receiving, at the fine-tuned generative model, computer programming code; This claim recites the additional element “receiving” which is merely an insignificant extra-solution activity such as gathering data, and does not integrate the judicial exception into a practical application (see MPEP § 2106.05(g)), analyzed further below in Step 2B as being well-understood, routine, and conventional.
via the fine-tuned generative mode; This limitation recites additional elements that merely recite instructions to implement an abstract idea on a generic computer. (See MPEP § 2106.05(f)).
Claim 8 additionally recites, “one or more processors; and one or more memories coupled with the one or more processors and storing processor-executable code that, when executed by the one or more processors, is configured to cause the apparatus to”; This limitation recites additional elements that merely recite instructions to implement an abstract idea on a generic computer. (See MPEP § 2106.05(f)).
Claim 15 additionally recites, “A non-transitory computer-readable medium having program code recorded thereon, the program code executed by one or more processors and comprising”; This limitation recites additional elements that merely recite instructions to implement an abstract idea on a generic computer. (See MPEP § 2106.05(f)).
Claim 15 additionally recites, “program code to”; This limitation recites additional elements that merely recite instructions to implement an abstract idea on a generic computer. (See MPEP § 2106.05(f)).
Step 2B Analysis:
The additional elements, considering them both individually and in combination, do not amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract ideas into a practical application, the additional elements which are insignificant extra-solution activity such as gathering data, are well-understood, routine, and conventional (see MPEP § 2106.05(d)(II) for court decisions recognizing that this activity is well-understood, routine, and conventional.). The claim is not patent eligible.
Regarding Claims 2, 9, and 16:
Step 2A Prong 1 Analysis:
See corresponding analysis regarding Claims 1, 8, and 15.
Step 2A Prong 2 Analysis:
The claim limitation recites, wherein the group of algorithms are Python algorithms; This limitation recites additional elements that indicate a field of use or technological environment in which to apply a judicial exception. The claim merely recites “Python algorithms” as a technological environment to perform the abstract ideas. (See MPEP § 2106.05(h)).
Step 2B Analysis:
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 ideas into a practical application, all the additional elements merely indicate a field of use or technological environment in which to apply a judicial exception. The claim is not patent eligible.
Regarding Claims 3, 10, and 17:
Step 2A Prong 1 Analysis:
See corresponding analysis regarding Claims 2, 9, and 16.
Step 2A Prong 2 Analysis:
The claim limitation recites, wherein the virtual machine is a Python virtual machine; This limitation recites additional elements that indicate a technological environment in which to apply a judicial exception. The claim merely recites “Python virtual machine” as a technological environment to perform the abstract ideas. (See MPEP § 2106.05(h)).
Step 2B Analysis:
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 ideas into a practical application, all the additional elements merely indicate a field of use or technological environment in which to apply a judicial exception. The claim is not patent eligible.
Regarding Claims 4, 11, and 18:
Step 2A Prong 1 Analysis:
See corresponding analysis regarding Claims 1, 8, and 15.
Step 2A Prong 2 Analysis:
The claim limitation recites, wherein the generative model is a large language model (LLM); This limitation recites additional elements that indicate a field of use in which to apply a judicial exception. The claim merely recites “large language model” as a field of use to perform the abstract ideas. (See MPEP § 2106.05(h)).
Step 2B Analysis:
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 ideas into a practical application, all the additional elements merely indicate a field of use or technological environment in which to apply a judicial exception. The claim is not patent eligible.
Regarding Claims 5, 12, and 19:
Step 2A Prong 1 Analysis:
The claim limitation recites, to generate sequences corresponding to the code trace; This limitation covers performance in the mind in the form of evaluation and judgement with the assistance of pen and paper. For example, a developer, with pen and paper, may create sequences of step-by-step descriptions on how code is traced at execution. Therefore, this limitation recites a mental process. (See MPEP § 2106.04(a)(2), subsection III).
Step 2A Prong 2 Analysis:
wherein the generative model is fine-tuned; This limitation recites additional elements that merely uses a generic computer or computer components as a tool to perform the abstract idea. The generative model is a generic computer merely being used to generate the mental process described above in Step 2A Prong 1. (See MPEP § 2106.05(f)).
Step 2B Analysis:
The additional elements, considering them both individually and in combination, do not amount to significantly more than the judicial exception. The claim is not patent eligible.
Regarding Claims 6, 13, and 20:
Step 2A Prong 1 Analysis:
See corresponding analysis regarding Claims 1, 8, and 15.
Step 2A Prong 2 Analysis:
The claim limitation recites, wherein each code trace traces the respective algorithm at a function level; This limitation recites additional elements that indicate a field of use in which to apply a judicial exception. The claim merely recites “at a function level” as a field of use to perform the abstract ideas. (See MPEP § 2106.05(h)).
Step 2B Analysis:
The additional elements, considering them both individually and in combination, do not amount to significantly more than the judicial exception. The claim is not patent eligible.
Regarding Claims 7 and 14:
Step 2A Prong 1 Analysis:
See corresponding analysis regarding Claims 1, 8, and 15.
Step 2A Prong 2 Analysis:
The claim limitation recites, wherein fine-tuned generative model interacts with an interpreter associated with the computer programming code; This claim recites the additional element “interacts with an interpreter” which is merely an insignificant extra-solution activity, being a post-solution to the generative model when receiving computer programming code. A computer in most practical cases interacts with an interpreter when processing data, such as when receiving the computer programming code, from a higher-level programming language abstraction to a lower-level abstraction of machine code in order to communicate with the computer’s hardware. Therefore, the claim limitation does not integrate the judicial exception into a practical application (see MPEP § 2106.05(g)), analyzed further below in Step 2B as being well-understood, routine, and conventional.
Step 2B Analysis:
The additional elements, considering them both individually and in combination, do not amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract ideas into a practical application, the additional elements which are insignificant extra-solution activity such as gathering data, are well-understood, routine, and conventional (see MPEP § 2106.05(d)(II) for court decisions recognizing that this activity is well-understood, routine, and conventional.). The claim is not patent eligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 5-6, 8-9, 12-13, 15-16, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Wang (U.S. Publication No. 20220044119 A1, hereinafter Wang) in view of Cser (U.S. Publication No. 20240320131 A1, hereinafter Cser).
Regarding Claim 1:
Wang discloses,
“A method comprising” (In paragraph [0008], “implementing a method including receiving a plurality of execution traces of a program”.);
“generating, ” (In paragraph [0058], “In step 220, a dataset can be generated from the programs. The dataset may comprise a plurality of execution traces for each of the computer programs. Each program may be run with different inputs, and an execution trace may be collected for each execution of each program”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”.);
(Examiner’s Note: Note the alternative term usage defined in the instant application, paragraph [0055] of the specification, “execution traces, which may also be referred to as code traces (hereinafter used interchangeably)”.
“[] in accordance with the group of code traces” (In paragraph [0059], “The dataset can also be divided into a training dataset … The training dataset may be used to train the deep learning model”. In paragraph [0058], “The dataset may comprise a plurality of execution traces”.);
“receiving, [], computer programming code” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language”. In paragraph [0090], “The learning model 642, in conjunction with the processor 620, may receive data including execution traces from a program or a plurality of programs”.
In paragraph [0105], “the dataset, including programs and/or execution traces, can be labeled with semantic labels”. In paragraph [0106], “one or more deep learning models may be trained on a training subset of the dataset comprising the labeled execution traces and/or the labeled programs”.).
Wang does not disclose however Cser discloses,
“via a virtual machine” (In paragraph [0054]);
“fine-tuning a generative model” (In paragraph [0005], “training the base model with an application-specific training dataset to provide a fine-tuned model for an application”. In paragraph [0057], “Provided herein is a technology … to train a machine learning model (e.g., a generative pretrained transformer deep learning model)”.);
“at the fine-tuned generative model” (In paragraph [0081], “sequences of user actions are input to the fine-tuned model”. In paragraph [0074], “manually scripted test cases describe sequential user actions performed”. In paragraph [0072], “convert the sequence of integers to a sequence of actions on a graphical user interface encoded in a programming language (e.g., a test script)”.);
“and generating, via the fine-tuned generative mode, one or more computer programming code statements corresponding to the computer programming code or simulate an expected output of the computer programming code” (In paragraph [0057], “and using 2400 the fine-tuned model to generate software application test cases”. In paragraph [0005], “methods further comprise generating executable code in a programming language to perform the software application test case”.
In paragraph [0094], “the technology comprises a runtime agent that simulates user actions on an application. The runtime agent is configured to request a predicted next step from the fine-tuned model at runtime and receives the predicted next step from the fine-tuned model. The runtime agent then evaluates the prediction by attempting to execute the predicted next test step”. In paragraph [0095], “the technology further comprises methods for generating program code (e.g., a test script) from the token sequence produced by the fine-tuned model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 2:
Wang discloses,
“wherein the group of algorithms are Python algorithms” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language. In this example, program 412 is written in Python”.).
Regarding Claim 5:
Wang discloses,
“wherein the [] is fine-tuned to generate sequences corresponding to the code trace” (In paragraph [0065], “the deep learning model can embed the vocabulary embeddings into a state embedding vector with a first recurrent neural network RNN1. The sequence of vocabulary tokens for each program state can be run through the first recurrent neural network … The variable u.sub.e_s_v.sub.n can represent the value of the n-th variable v.sub.n of the s-th program state in the e-th execution trace”. In paragraph [0067], “the deep learning model can compute a forward sequence of program states … Given the state embeddings h of an execution e (h.sub.e_1 to h.sub.e_n)”.).
Wang does not disclose however Cser discloses,
“generative model” (In paragraph [0065], “The GPT model uses the self-attention mechanism to focus on specific parts of the input sequence to generate the output sequence”. In paragraph [0058], “a generative pretrained transformer (GPT) model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 6:
Wang discloses,
“wherein each code trace traces the respective algorithm at a function level” (In paragraph [0042], “An execution trace may be a record of the execution of a computer program, and may be broken into a plurality of tokens (e.g., values of variables in the program)”. In paragraph [0026], “An execution trace may, for example, record each line of source code as it is executed”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”.).
Regarding Claim 8:
Wang discloses,
“An apparatus, comprising: one or more processors; and one or more memories coupled with the one or more processors and storing processor-executable code that, when executed by the one or more processors, is configured to cause the apparatus to” (In paragraph [0008], “the disclosure includes a computer comprising a processor and a computer-readable medium comprising code, executable by the processor, for implementing a method including receiving a plurality of execution traces of a program”.);
“generate, ” (In paragraph [0058], “In step 220, a dataset can be generated from the programs. The dataset may comprise a plurality of execution traces for each of the computer programs. Each program may be run with different inputs, and an execution trace may be collected for each execution of each program”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”.);
(Examiner’s Note: Note the alternative term usage defined in the instant application, paragraph [0055] of the specification, “execution traces, which may also be referred to as code traces (hereinafter used interchangeably)”.
“[] in accordance with the group of code traces” (In paragraph [0059], “The dataset can also be divided into a training dataset … The training dataset may be used to train the deep learning model”. In paragraph [0058], “The dataset may comprise a plurality of execution traces”.);
“receive, [], computer programming code” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language”. In paragraph [0090], “The learning model 642, in conjunction with the processor 620, may receive data including execution traces from a program or a plurality of programs”.
In paragraph [0105], “the dataset, including programs and/or execution traces, can be labeled with semantic labels”. In paragraph [0106], “one or more deep learning models may be trained on a training subset of the dataset comprising the labeled execution traces and/or the labeled programs”.).
Wang does not disclose however Cser discloses,
“via a virtual machine” (In paragraph [0054]);
“fine-tune a generative model” (In paragraph [0005], “training the base model with an application-specific training dataset to provide a fine-tuned model for an application”. In paragraph [0057], “Provided herein is a technology … to train a machine learning model (e.g., a generative pretrained transformer deep learning model)”.);
“at the fine-tuned generative model” (In paragraph [0081], “sequences of user actions are input to the fine-tuned model”. In paragraph [0074], “manually scripted test cases describe sequential user actions performed”. In paragraph [0072], “convert the sequence of integers to a sequence of actions on a graphical user interface encoded in a programming language (e.g., a test script)”.);
“and generate, via the fine-tuned generative mode, one or more computer programming code statements corresponding to the computer programming code or simulate an expected output of the computer programming code” (In paragraph [0057], “and using 2400 the fine-tuned model to generate software application test cases”. In paragraph [0005], “methods further comprise generating executable code in a programming language to perform the software application test case”.
In paragraph [0094], “the technology comprises a runtime agent that simulates user actions on an application. The runtime agent is configured to request a predicted next step from the fine-tuned model at runtime and receives the predicted next step from the fine-tuned model. The runtime agent then evaluates the prediction by attempting to execute the predicted next test step”. In paragraph [0095], “the technology further comprises methods for generating program code (e.g., a test script) from the token sequence produced by the fine-tuned model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 9:
Wang discloses,
“wherein the group of algorithms are Python algorithms” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language. In this example, program 412 is written in Python”.).
Regarding Claim 12:
Wang discloses,
“wherein the [] is fine-tuned to generate sequences corresponding to the code trace” (In paragraph [0065], “the deep learning model can embed the vocabulary embeddings into a state embedding vector with a first recurrent neural network RNN1. The sequence of vocabulary tokens for each program state can be run through the first recurrent neural network … The variable u.sub.e_s_v.sub.n can represent the value of the n-th variable v.sub.n of the s-th program state in the e-th execution trace”. In paragraph [0067], “the deep learning model can compute a forward sequence of program states … Given the state embeddings h of an execution e (h.sub.e_1 to h.sub.e_n)”.).
Wang does not disclose however Cser discloses,
“generative model” (In paragraph [0065], “The GPT model uses the self-attention mechanism to focus on specific parts of the input sequence to generate the output sequence”. In paragraph [0058], “a generative pretrained transformer (GPT) model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 13:
Wang discloses,
“wherein each code trace traces the respective algorithm at a function level” (In paragraph [0042], “An execution trace may be a record of the execution of a computer program, and may be broken into a plurality of tokens (e.g., values of variables in the program)”. In paragraph [0026], “An execution trace may, for example, record each line of source code as it is executed”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”.).
Regarding Claim 15:
Wang discloses,
“A non-transitory computer-readable medium having program code recorded thereon, the program code executed by one or more processors and comprising” (In paragraph [0008], “the disclosure includes … a computer-readable medium comprising code, executable by the processor, for implementing a method including receiving a plurality of execution traces of a program”.);
“program code to generate, ” (In paragraph [0058], “In step 220, a dataset can be generated from the programs. The dataset may comprise a plurality of execution traces for each of the computer programs. Each program may be run with different inputs, and an execution trace may be collected for each execution of each program”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”. In paragraph [0008], “the disclosure includes … a computer-readable medium comprising code, executable by the processor, for implementing a method”.);
(Examiner’s Note: Note the alternative term usage defined in the instant application, paragraph [0055] of the specification, “execution traces, which may also be referred to as code traces (hereinafter used interchangeably)”.
“program code to [] in accordance with the group of code traces” (In paragraph [0059], “The dataset can also be divided into a training dataset … The training dataset may be used to train the deep learning model”. In paragraph [0058], “The dataset may comprise a plurality of execution traces”. In paragraph [0008], “the disclosure includes … a computer-readable medium comprising code, executable by the processor, for implementing a method”.);
“program code to receive, [], computer programming code” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language”. In paragraph [0090], “The learning model 642, in conjunction with the processor 620, may receive data including execution traces from a program or a plurality of programs”.
In paragraph [0105], “the dataset, including programs and/or execution traces, can be labeled with semantic labels”. In paragraph [0106], “one or more deep learning models may be trained on a training subset of the dataset comprising the labeled execution traces and/or the labeled programs”. ).
Wang does not disclose however Cser discloses,
“via a virtual machine” (In paragraph [0054]);
“fine-tune a generative model” (In paragraph [0005], “training the base model with an application-specific training dataset to provide a fine-tuned model for an application”. In paragraph [0057], “Provided herein is a technology … to train a machine learning model (e.g., a generative pretrained transformer deep learning model)”.);
“at the fine-tuned generative model” (In paragraph [0081], “sequences of user actions are input to the fine-tuned model”. In paragraph [0074], “manually scripted test cases describe sequential user actions performed”. In paragraph [0072], “convert the sequence of integers to a sequence of actions on a graphical user interface encoded in a programming language (e.g., a test script)”.);
“and program code to generate, via the fine-tuned generative mode, one or more computer programming code statements corresponding to the computer programming code or simulate an expected output of the computer programming code” (In paragraph [0057], “and using 2400 the fine-tuned model to generate software application test cases”. In paragraph [0005], “methods further comprise generating executable code in a programming language to perform the software application test case”.
In paragraph [0094], “the technology comprises a runtime agent that simulates user actions on an application. The runtime agent is configured to request a predicted next step from the fine-tuned model at runtime and receives the predicted next step from the fine-tuned model. The runtime agent then evaluates the prediction by attempting to execute the predicted next test step”. In paragraph [0095], “the technology further comprises methods for generating program code (e.g., a test script) from the token sequence produced by the fine-tuned model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 16:
Wang discloses,
“wherein the group of algorithms are Python algorithms” (In paragraph [0076], “Program 412 may be any computer program, written in an appropriate computer programming language. In this example, program 412 is written in Python”.).
Regarding Claim 19:
Wang discloses,
“wherein the [] is fine-tuned to generate sequences corresponding to the code trace” (In paragraph [0065], “the deep learning model can embed the vocabulary embeddings into a state embedding vector with a first recurrent neural network RNN1. The sequence of vocabulary tokens for each program state can be run through the first recurrent neural network … The variable u.sub.e_s_v.sub.n can represent the value of the n-th variable v.sub.n of the s-th program state in the e-th execution trace”. In paragraph [0067], “the deep learning model can compute a forward sequence of program states … Given the state embeddings h of an execution e (h.sub.e_1 to h.sub.e_n)”.).
Wang does not disclose however Cser discloses,
“generative model” (In paragraph [0065], “The GPT model uses the self-attention mechanism to focus on specific parts of the input sequence to generate the output sequence”. In paragraph [0058], “a generative pretrained transformer (GPT) model”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Regarding Claim 20:
Wang discloses,
“wherein each code trace traces the respective algorithm at a function level” (In paragraph [0042], “An execution trace may be a record of the execution of a computer program, and may be broken into a plurality of tokens (e.g., values of variables in the program)”. In paragraph [0026], “An execution trace may, for example, record each line of source code as it is executed”. In paragraph [0059], “each execution trace may be labeled with the semantics of the program. Semantic labels may include the program functionality”.).
Claims 3, 7, 10, 14 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Wang in view of Cser as applied to Claims 1, 2, 8, 9, and 16 above, and further in view of Kim et al. (U.S. Publication No. 20250315693 A1, hereinafter Kim).
Regarding Claim 3:
Wang as modified does not disclose however Kim discloses,
“wherein the virtual machine is a Python virtual machine” (In paragraph [0060], “Referring to FIG. 4, the Python interpreter reads source code and converts it into bytecode, which is an intermediate form understandable by Python virtual machine (PVM). Although Python is an interpreted language, it actually compiles the source code into bytecode and interprets and executes this bytecode in the PVM”.)
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a Python virtual machine in Kim; motivated by the common goal to improve a model’s performance, such as with accuracy (Kim [0041, 0004]).
Regarding Claim 10:
Wang as modified does not disclose however Kim discloses,
“wherein the virtual machine is a Python virtual machine” (In paragraph [0060], “Referring to FIG. 4, the Python interpreter reads source code and converts it into bytecode, which is an intermediate form understandable by Python virtual machine (PVM). Although Python is an interpreted language, it actually compiles the source code into bytecode and interprets and executes this bytecode in the PVM”.)
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a Python virtual machine in Kim; motivated by the common goal to improve a model’s performance, such as with accuracy (Kim [0041, 0004]).
Regarding Claim 7:
Wang does not disclose however Cser discloses,
“wherein fine-tuned generative model interacts with [] associated with the computer programming code” (In paragraph [0039], “Thus, the methods and apparatus of the embodiments, or certain aspects or portions thereof, may take the form of program code (e.g., instructions) … wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the embodiments … the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations”.
In paragraph [0058], “As described herein, embodiments of the server may comprise a generative pretrained transformer (GPT) model”. In paragraph [0077], “The process of fine-tuning comprises providing the base model (e.g., the GPT model trained with base user interaction training data)”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Wang as modified does not disclose however Kim discloses,
“an interpreter” (In paragraph [0060], “Referring to FIG. 4, the Python interpreter reads source code and converts it into bytecode”. In paragraph [0065], “model analyzer 51 generates a graph tree by analyzing and tracing the structure of the model based on the model's bytecode”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of an interpreter in Kim; motivated by the common goal to improve a model’s performance, such as with accuracy (Kim [0041, 0004]).
Regarding Claim 14:
Wang does not disclose however Cser discloses,
“wherein fine-tuned generative model interacts with [] associated with the computer programming code” (In paragraph [0039], “Thus, the methods and apparatus of the embodiments, or certain aspects or portions thereof, may take the form of program code (e.g., instructions) … wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the embodiments … the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations”.
In paragraph [0058], “As described herein, embodiments of the server may comprise a generative pretrained transformer (GPT) model”. In paragraph [0077], “The process of fine-tuning comprises providing the base model (e.g., the GPT model trained with base user interaction training data)”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to modify Wang by adopting the teachings of Cser; motivated by the common goal to improve the performance of a generative model, such as with the technique of fine-tuning (Cser [0081]).
Wang as modified does not disclose however Kim discloses,
“an interpreter” (In paragraph [0060], “Referring to FIG. 4, the Python interpreter reads source code and converts it into bytecode”. In paragraph [0065], “model analyzer 51 generates a graph tree by analyzing and tracing the structure of the model based on the model's bytecode”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of an interpreter in Kim; motivated by the common goal to improve a model’s performance, such as with accuracy (Kim [0041, 0004]).
Regarding Claim 17:
Wang as modified does not disclose however Kim discloses,
“wherein the virtual machine is a Python virtual machine” (In paragraph [0060], “Referring to FIG. 4, the Python interpreter reads source code and converts it into bytecode, which is an intermediate form understandable by Python virtual machine (PVM). Although Python is an interpreted language, it actually compiles the source code into bytecode and interprets and executes this bytecode in the PVM”.)
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a Python virtual machine in Kim; motivated by the common goal to improve a model’s performance, such as with accuracy (Kim [0041, 0004]).
Claims 4, 11, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Wang in view of Cser as applied to Claims 1, 8, and 15 above, and further in view of Rochlitzer et al. (U.S. Publication No. 20250278428 A1, hereinafter Rochlitzer).
Regarding Claim 4:
Wang as modified does not disclose however Rochlitzer discloses,
“wherein the generative model is a large language model (LLM)” (In paragraph [0049], “The component 1940 may include generative LLMs as well as LLMs that compare process elements”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a LLM in Rochlitzer; motivated by the common goal to improve a developer’s experience by suggesting improvements with machine learning, through the technique of collecting execution traces (Rochlitzer [0026, 0029]).
Regarding Claim 11:
Wang as modified does not disclose however Rochlitzer discloses,
“wherein the generative model is a large language model (LLM)” (In paragraph [0049], “The component 1940 may include generative LLMs as well as LLMs that compare process elements”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a LLM in Rochlitzer; motivated by the common goal to improve a developer’s experience by suggesting improvements with machine learning, through the technique of collecting execution traces (Rochlitzer [0026, 0029]).
Regarding Claim 18:
Wang as modified does not disclose however Rochlitzer discloses,
“wherein the generative model is a large language model (LLM)” (In paragraph [0049], “The component 1940 may include generative LLMs as well as LLMs that compare process elements”.).
Therefore, it would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to further modify Wang by adopting the teaching of a LLM in Rochlitzer; motivated by the common goal to improve a developer’s experience by suggesting improvements with machine learning, through the technique of collecting execution traces (Rochlitzer [0026, 0029]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Beza D Nigatu whose telephone number is (571)272-9643. The examiner can normally be reached Monday - Friday 7:30am-3:30pm.
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, Hyung Sough can be reached at (571) 272-6799. 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.
/BEZA D NIGATU/Examiner, Art Unit 2192
/S. Sough/SPE, Art Unit 2192
Prosecution Insights