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 .
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
18/678,628 Claim
US 20250371383 A1 Claim
1
A method for managing a generative inference model, the method comprising:
1
A method for managing a generative inference model, the method comprising:
1
obtaining an inference generated using the generative inference model and ingest data;
1
obtaining an inference generated using the generative inference model and ingest data;
1
obtaining a summarization data package for the inference using at least a summarization model;
1
analyzing the inference using a schema to obtain a set of concepts displayed by the inference;
1
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works;
1
obtaining a structured representation of the inference based on the set of concepts; performing a comparison process using the structured representation of the inference and structured representations of reference works to obtain a level of similarity between the inference and the reference works;
1
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and
1
making a determination regarding whether the inference is acceptable based on the level of similarity and a similarity threshold; and
1
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and
1
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and
1
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works.
1
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of the level of similarity between the inference and the reference works.
2
The method of claim 1, wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object.
3
The method of claim 2, wherein the foundation model is further adapted to limit a quantity of information added to the summarization data package for the inference.
4
The method of claim 3, wherein the foundation model is adapted to use a summary schema to generate the summarization data package for the inference, the summary schema discriminating the subset of the information from other information from the source data object.
5
The method of claim 4, wherein the summary schema discriminates conceptual information from the source data object from contextual information from the source data object.
6
The method of claim 1, wherein the generative inference model is trained to generate human interpretable text when prompted using the ingest data, the human interpretable text being responsive to a request indicated by the ingest data.
2
The method of claim 1, wherein the generative inference model is trained to generate human interpretable text when prompted using the ingest data, the human interpretable text being responsive to a request indicated by the ingest data.
7
The method of claim 6, further comprising: obtaining a structured representation for the inference based on the summarization data package for the inference using a structured representation schema, wherein the structured representation for the inference is used during the obtaining of the of the levels of similarity as a basis of comparison for the inference to the reference works.
1
obtaining a structured representation of the inference based on the set of concepts; performing a comparison process using the structured representation of the inference and structured representations of reference works to obtain a level of similarity between the inference and the reference works;
8
The method of claim 7, wherein the structured representation schema is adapted to facilitate identification of at least one of: a location indicated by the inference; a character indicated by the inference; an object indicated by the inference; and a law of nature indicated by the inference.
3
The method of claim 1, wherein the schema is adapted to facilitate identification of at least one of: a location indicated by the inference; a character indicated by the inference; an object indicated by the inference; and a law of nature indicated by the inference.
9
The method of claim 1, wherein obtaining the summarization data package for the inference comprises: obtaining an objective, the objective indicating at least one constraint for adding information to the summarization data package for the inference; and prompting, using at least the objective, the summarization model to generate the summarization data package for the inference.
10
The method of claim 1, further comprising: prior to obtaining the levels of similarity: for a portion of the reference works: obtaining a summarization data package for the portion of the reference works using at least the summarization model; and obtaining a structured representation for the portion of the reference works based on the summarization data package for the portion of the reference works using a structured representation schema.
6
The method of claim 1, further comprising: prior to performing the comparison process: for a portion of the reference works: analyzing the portion of the reference works to obtain a set of concepts associated with the portion of the reference works; and obtaining a structured representation of the portion of the reference works based on the set of concepts associated with the portion of the reference works.
11
The method of claim 7, wherein the structured representation for the inference comprises a graph-structured data model that specifies relationships between elements of the summarization data package for the inference, the graph-structured data model comprising nodes and edges, and the edges being based on the relationships between the elements associated with the edges.
4
The method of claim 1, wherein the structured representation of the inference comprises a graph-structured data model that specifies relationships between concepts of the set of concepts, the graph-structured data model comprising nodes and edges, and the edges being based on the relationships between the concepts associated with the edges.
12
The method of claim 11, wherein obtaining the levels of similarity comprises: performing a sub-graph analysis of the structured representation for the inference with respect to portions of structured representations for the reference works to identify whether a portion of the structured representation for the inference substantially matches one of the portions of the structured representations for the reference works.
5
The method of claim 1, wherein performing the comparison process comprises: performing a sub-graph analysis of the structured representation of the inference with respect to portions of the structured representations of the reference works to identify whether a portion of the structured representation of the inference substantially matches one of the portions of the structured representations of the reference works.
13
The method of claim 1, wherein the levels of similarity indicate likelihoods that the inference plagiarizes the reference works.
7
The method of claim 1, wherein the level of similarity indicates a likelihood that the inference plagiarizes the reference works.
14
The method of claim 1, wherein the action set comprises obtaining a description of the similarities between the inference and the at least one of the reference works.
8
The method of claim 1, wherein the action set comprises obtaining a description of similarities between the inference and the reference works.
15
The method of claim 1, wherein the action set comprises preventing provision of the inference to the downstream consumer.
9
The method of claim 1, wherein the action set comprises preventing provision of the inference to the downstream consumer.
16
The method of claim 1, wherein the action set comprises at least one action that, when performed, modifies operation and/or use of the generative inference model to reduce a likelihood that a future inference generated using the generative inference model and the ingest data plagiarizes the reference works.
10
The method of claim 1, wherein the action set comprises at least one action that, when performed, modifies operation and/or use of the generative inference model to reduce a likelihood that a future inference generated using the generative inference model and the ingest data plagiarizes the reference works.
17
A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing a generative inference model, the operations comprising:
11
A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing a generative inference model, the operations comprising:
17
obtaining an inference generated using the generative inference model and ingest data;
11
obtaining an inference generated using the generative inference model and ingest data;
17
obtaining a summarization data package for the inference using at least a summarization model;
11
analyzing the inference using a schema to obtain a set of concepts displayed by the inference;
17
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works;
11
obtaining a structured representation of the inference based on the set of concepts; performing a comparison process using the structured representation of the inference and structured representations of reference works to obtain a level of similarity between the inference and the reference works;
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and
11
making a determination regarding whether the inference is acceptable based on the level of similarity and a similarity threshold; and
17
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and
11
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and
17
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works.
11
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of the level of similarity between the inference and the reference works.
18
The non-transitory machine-readable medium of claim 17, wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object.
19
A data processing system, comprising: a processor; and a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing a generative inference model, the operations comprising:
16
A data processing system, comprising: a processor; and a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing a generative inference model, the operations comprising:
19
obtaining an inference generated using the generative inference model and ingest data;
16
obtaining an inference generated using the generative inference model and ingest data,
19
obtaining a summarization data package for the inference using at least a summarization model;
16
analyzing the inference using a schema to obtain a set of concepts displayed by the inference,
19
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works;
16
obtaining a structured representation of the inference based on the set of concepts, performing a comparison process using the structured representation of the inference and structured representations of reference works to obtain a level of similarity between the inference and the reference works,
19
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and
16
making a determination regarding whether the inference is acceptable based on the level of similarity and a similarity threshold, and
Although the conflicting claims are not identical, they are not patentably distinct from each other. The claims of the instant application and the claims of copending Application No. 18/678,530 are directed to the same method of managing a generative inference model: obtaining an inference generated using the generative inference model and ingest data, deriving a reduced, concept-level representation of the inference, obtaining a level of similarity between the inference and reference works using that representation, comparing the level of similarity to a similarity threshold to determine acceptability, and either providing the inference to a downstream consumer or initiating an action set. The instant claims differ only in reciting that the reduced representation (a "summarization data package") is obtained "using at least a summarization model," whereas the US 20250371383 A1 claims recite "analyzing the inference using a schema to obtain a set of concepts." This difference does not render the claims patentably distinct because, under the broadest reasonable interpretation consistent with the instant specification, the summarization data package may consist of concept data alone (see instant specification, [0097], [0107]) and the summarization model encompasses any inference model adapted to extract a subset of information from a source data object (pargagraphs [0085]–[0086]). Performing the claimed schema-based concept extraction using a trained model, such as a foundation model, was a well-known and conventional implementation choice at the time of filing, as evidenced by Kryscinski (US 20220277135 A1), which teaches an encoder-based summarization model that extracts relevant portions of a source document in accordance with a query and synthesizes them into a summary (pargagraphs [0016]–[0019]). Accordingly, it would have been obvious to one of ordinary skill in the art to implement the concept-extraction step of the US 20250371383 A1 claims using a summarization model as claimed, and the instant claims are therefore an obvious variant of the US 20250371383 A1 claims.
Claim Objections
Claim 4 objected to because of the following informalities: the claim recites "the subset of the information" and in claim 2 it is introduced as "a subset of information", should be corrected to "the subset of information". Appropriate correction is required.
Claim 7 objected to because of the following informalities: the claim recites "of the of the" should be corrected to "of the". Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
MPEP 2106 (III) sets out steps for evaluating whether a claim is drawn to patent-eligible subject matter. The analysis of claims 1-20, in accordance with these steps, follows.
Step 1 Analysis:
Claims 1-16 are directed to method (processes). Claims 19-20 are directed to a data processing system (machine). Claims 17-18 are directed to a non-transitory machine-readable medium (article of manufacture). Therefore, claims 1-20 fall into one of four statutory categories (i.e., process, machine, article of manufacture).
As to claim 1,
Step 2A Prong 1: this claim recites the following abstract ideas:
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (the limitation describes comparing the summarized content against reference works and judging how similar it is to each, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; (the limitation describes comparing the levels of similarity to a threshold to decide acceptability, which is a mental process implemented in the human mind.)
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (the limitation describes deciding, when the result is judged unacceptable, to take a set of responsive actions to address the similarities, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
obtaining an inference generated using the generative inference model and ingest data; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining a summarization data package for the inference ...; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
using at least a summarization model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; (this limitation describes data transmission, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 17,
Step 2A Prong 1: this claim recites the following abstract ideas:
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (the limitation describes comparing the summarized content against reference works and judging how similar it is to each, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; (the limitation describes comparing the levels of similarity to a threshold to decide acceptability, which is a mental process implemented in the human mind.)
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (the limitation describes deciding, when the result is judged unacceptable, to take a set of responsive actions to address the similarities, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing a generative inference model; (This limitation is directed to mere instruction to store the abstract idea on a generic memory and apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining an inference generated using the generative inference model and ingest data; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining a summarization data package for the inference ...; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
using at least a summarization model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; (this limitation describes data transmission, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 19,
Step 2A Prong 1: this claim recites the following abstract ideas:
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (the limitation describes comparing the summarized content against reference works and judging how similar it is to each, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; (the limitation describes comparing the levels of similarity to a threshold to decide acceptability, which is a mental process implemented in the human mind.)
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (the limitation describes deciding, when the result is judged unacceptable, to take a set of responsive actions to address the similarities, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
a processor; and a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing a generative inference model; (This limitation is directed to mere instruction to store the abstract idea on a generic memory and apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining an inference generated using the generative inference model and ingest data; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining a summarization data package for the inference ...; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
using at least a summarization model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; (this limitation describes data transmission, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claims 2, 18, and 20,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claims depend on claims 1, 17, and 19.
Step 2A Prong 2 and 2B: those claims recited the following additional elements:
wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 3,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 2.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the foundation model is further adapted to limit a quantity of information added to the summarization data package for the inference. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 4,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 3.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the foundation model is adapted to use a summary schema to generate the summarization data package for the inference, the summary schema discriminating the subset of the information from other information from the source data object. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 5,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein the summary schema discriminates conceptual information from the source data object from contextual information from the source data object. (the limitation describes the content of the information being discriminated, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 6,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the generative inference model is trained to generate human interpretable text when prompted using the ingest data, the human interpretable text being responsive to a request indicated by the ingest data. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 7,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 6.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
obtaining a structured representation for the inference based on the summarization data package for the inference using a structured representation schema, wherein the structured representation for the inference is used during the obtaining of the of the levels of similarity as a basis of comparison for the inference to the reference works. (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 8,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein the structured representation schema is adapted to facilitate identification of at least one of: a location indicated by the inference; a character indicated by the inference; an object indicated by the inference; and a law of nature indicated by the inference. (the limitation describes the content of the information being identified, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 9,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
obtaining an objective, the objective indicating at least one constraint for adding information to the summarization data package for the inference; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
prompting, using at least the objective, the summarization model to generate the summarization data package for the inference. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 10,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
prior to obtaining the levels of similarity: for a portion of the reference works: obtaining a summarization data package for the portion of the reference works ...; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
using at least the summarization model; (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
obtaining a structured representation for the portion of the reference works based on the summarization data package for the portion of the reference works using a structured representation schema. (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 11,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein the structured representation for the inference comprises a graph-structured data model that specifies relationships between elements of the summarization data package for the inference, the graph-structured data model comprising nodes and edges, and the edges being based on the relationships between the elements associated with the edges. (the limitation describes the content and form of the information used as the basis of comparison, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 12,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein obtaining the levels of similarity comprises: performing a sub-graph analysis of the structured representation for the inference with respect to portions of structured representations for the reference works to identify whether a portion of the structured representation for the inference substantially matches one of the portions of the structured representations for the reference works. (the limitation describes comparing portions of one representation against portions of other representations to identify matching portions, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 13,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein the levels of similarity indicate likelihoods that the inference plagiarizes the reference works. (the limitation describes the content of the information being evaluated, which merely specifies the type of data evaluated and is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 14,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the action set comprises obtaining a description of the similarities between the inference and the at least one of the reference works. (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 15,
Step 2A Prong 1: this claim recites the following abstract ideas:
wherein the action set comprises preventing provision of the inference to the downstream consumer. (the limitation describes deciding to withhold the result rather than provide it, which is an evaluation and judgment activity that can be performed as a mental process in the human mind.)
Step 2A Prong 2 and 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) 1.), failing step 2A prong 2. The claim is ineligible.
As to claim 16,
Step 2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on claim 1.
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
wherein the action set comprises at least one action that, when performed, modifies operation and/or use of the generative inference model to reduce a likelihood that a future inference generated using the generative inference model and the ingest data plagiarizes the reference works. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
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.
Claim(s) 1-6, 9, and 14-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Padgett et al. (US 20240160902 A1) in view of Kryscinski et al. (US 20220277135 A1).
As to claim 1, Padgett teaches a method for managing a generative inference model, the method comprising: (see Padgett paragraph [0062] "The system 100 includes a generative AI model 102, a similarity-assessment layer 104, and a repository of pre-existing content 106. The generative AI model 102 may be an unsupervised or semi-supervised machine learning algorithm that has been trained using a set of training data content.")
obtaining an inference generated using the generative inference model and ingest data; (see Padgett paragraph [0062] "The generative AI model 102 is configured to take an input prompt, typically in text form but may also possibly include images or other media inputs. The model 102 creates an output related to the input prompt.", and see Padgett paragraph [0076] "At operation 206, the generative AI model generates a result using the prompt as an input.")
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (see Padgett paragraph [0064] "The similarity-assessment layer 104 may compare the output to each item in the repository of pre-existing content 106 in turn. With each comparison the layer 104 may calculate a similarity measure between the output and the item.", and see Padgett paragraph [0077] "In some implementations, both the result and the item from the repository are processed through a simplifying filter and the filtered results are then compared to assess the distance metric.", and see Padgett paragraph [0113] "In some cases, the repository 406 contains the original training data set, such that the filtering operation excludes outputs that too closely resemble the training data, e.g. instances of rote learning or close-to rote learning.")
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and (see Padgett paragraph [0083] "In this example the similarity measure is compared to a threshold value in operation 210. If the index exceeds the threshold, then it is deemed "too similar" to one of the items from the repository. If below the threshold, then it is sufficiently different.")
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and (see Padgett paragraph [0083] "However similarity is assessed, if the result is not "too similar", then it is output at operation 214. The output of the result may include transmitting the result to a user device for display and/or storage on the user device in some cases.")
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (see Padgett paragraph [0084] "If, however, the result is determined to be "too similar" based on the similarity measure in operation 210, then the system adjusts the input to the AI model in operation 212 and returns to re-run the model so as to generate and evaluate new results.")
Padgett does not explicitly teach "obtaining a summarization data package for the inference using at least a summarization model"
However, Kryscinski teaches
obtaining a summarization data package for the inference using at least a summarization model; (see Kryscinski paragraph [0016] "Embodiments described herein provide a two-step QFS model, which includes an extractor model to extract parts of the source document relevant to the input query, and an abstractor model to synthesize the extracted segments into a final summary.", and see Kryscinski paragraph [0045] "At step 418, an abstractor model (e.g., 180 in FIG. 1B) may synthesize the extracted at least one part into a final summary of the testing source document.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett to include obtaining a summarization data package for the inference using at least a summarization model, as taught by Kryscinski, in order to produce condensed summaries focused on the information of interest, thereby enabling faster and more focused comparison of lengthy generated outputs against the repository of pre-existing content, with predictable results and a reasonable expectation of success, see Kryscinski paragraph [0015].
As to claim 2, Padgett as modified by Kryscinski teaches the method of claim 1, wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object. (see Kryscinski paragraph [0025] "FIG. 1B shows an example structure of the two-step QFS model that consists of an extractor model 160, which extracts parts of the source document 112 relevant to the input query 114, and an abstractor model 180, which synthesizes the extracted segments into a final summary 185 … In one embodiment, the abstractor model 180 may be a BART-large model.", and see Kryscinski paragraph [0053] "Model weights were initialized from pre-trained checkpoints available through the Huggingface Model Hub.")
As to claim 3, Padgett as modified by Kryscinski teaches the method of claim 2, wherein the foundation model is further adapted to limit a quantity of information added to the summarization data package for the inference. (see Kryscinski paragraph [0016] "Therefore, the relevance model can be used as a score-and-rank extractor model, which first score each source passage for relevance to the query and then rank the passages in descending order of relevance, with the concatenated and truncated results passed to the abstractor for synthesizing the final summary.")
As to claim 4, Padgett as modified by Kryscinski teaches the method of claim 3, wherein the foundation model is adapted to use a summary schema to generate the summarization data package for the inference, the summary schema discriminating the subset of the information from other information from the source data object. (see Kryscinski paragraph [0016] "This model is trained to predict the proxy for relevance (ROUGE) overlap between a given passage and the reference summary, using only the passage and query as input. Therefore, the relevance model can be used as a score-and-rank extractor model, which first score each source passage for relevance to the query and then rank the passages in descending order of relevance")
As to claim 5, Padgett as modified by Kryscinski teaches the method of claim 4, wherein the summary schema discriminates conceptual information from the source data object from contextual information from the source data object. (see Kryscinski paragraph [0015] "Query-focused summarization (QFS) is configured to produce summaries that answer particular questions of interest, enabling greater user control and personalization. Each source document can be associated with multiple unique queries inquiring about different information from that document.", and see Kryscinski paragraph [0016] "Therefore, the relevance model can be used as a score-and-rank extractor model, which first score each source passage for relevance to the query and then rank the passages in descending order of relevance, with the concatenated and truncated results passed to the abstractor for synthesizing the final summary.")
As to claim 6, Padgett as modified by Kryscinski teaches the method of claim 1, wherein the generative inference model is trained to generate human interpretable text when prompted using the ingest data, the human interpretable text being responsive to a request indicated by the ingest data. (see Padgett paragraph [0059] "Inputs to an LLM may be referred to as a prompt, which is a natural language input that includes instructions to the LLM to generate a desired output.", and see Padgett paragraph [0076] "The result may be a block of text or prose.")
As to claim 9, Padgett as modified by Kryscinski teaches the method of claim 1, wherein obtaining the summarization data package for the inference comprises:
obtaining an objective, the objective indicating at least one constraint for adding information to the summarization data package for the inference; and (see Kryscinski paragraph [0035] "At step 402, a source document (e.g., 102 in FIG. 1A), an input query (e.g., 104 in FIG. 1A) posing a question on a content of the source document and a reference summary (e.g., 106 in FIG. 1) of the source document may be received, via a communication interface (e.g., 315 in FIG. 3).")
prompting, using at least the objective, the summarization model to generate the summarization data package for the inference. (see Kryscinski paragraph [0022] "In one embodiment, the encoder model 120 is a single encoder model that concatenates a query 104 and a source passage 103a-n as input to the scoring function that produces the similarity score (e.g., the relevance value 133).", and see Kryscinski paragraph [0049] "At step 506, an encoder (e.g., 220 in FIG. 2) may encode each overlapping segment separately appended with the input query into a respective encoding.")
As to claim 14, Padgett as modified by Kryscinski teaches the method of claim 1, wherein the action set comprises obtaining a description of the similarities between the inference and the at least one of the reference works. (see Padgett paragraph [0064] "As it progresses through the comparisons between the output and the items in the repository of pre-existing content 106, the layer 104 may retain only the highest similarity measure. In some cases, it stores the highest similarity measure and at least an identifier of the corresponding item from the repository of pre-existing content 106 in memory.")
As to claim 15, Padgett as modified by Kryscinski teaches the method of claim 1, wherein the action set comprises preventing provision of the inference to the downstream consumer. (see Padgett paragraph [0096] "It may discard any results having similarity measures above the threshold value as candidate results on the basis that they are too similar to the pre-existing content to be considered as options.", and see Padgett paragraph [0112] "The similarity-assessment layer 404 evaluates the similarity measure of each output and filters the set of outputs to exclude one or more outputs based on their similarity measures so as to produce a set of filtered outputs 408.")
As to claim 16, Padgett as modified by Kryscinski teaches the method of claim 1, wherein the action set comprises at least one action that, when performed, modifies operation and/or use of the generative inference model to reduce a likelihood that a future inference generated using the generative inference model and the ingest data plagiarizes the reference works. (see Padgett paragraph [0087] "In addition to, or as an alternative to, modifying the prompt, the system may inject noise into the generative process at a point after the input and prior to the result. For example, in a cases where the AI model operates in a multi-stage manner to generate interim results and refine them in an iterative process, the system may be configured to inject noise at one of the stages to alter an interim result, thereby steering the AI model away from the original result that had been found "too similar".")
As to claim 17, Padgett teaches a non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing a generative inference model, the operations comprising: (see Padgett paragraph [0133] "The memory 610 allows data to be stored and retrieved. The memory 610 may include, for example, random access memory, read-only memory, and persistent storage. Persistent storage may be, for example, flash memory, a solid-state drive or the like. Read-only memory and persistent storage are a computer-readable medium.", and see Padgett paragraph [0137] "Software comprising instructions is executed by the processor 600 from a computer-readable medium. For example, software may be loaded into random-access memory from persistent storage of memory 610.")
obtaining an inference generated using the generative inference model and ingest data; (see Padgett paragraph [0062] "The generative AI model 102 is configured to take an input prompt, typically in text form but may also possibly include images or other media inputs. The model 102 creates an output related to the input prompt.", and see Padgett paragraph [0076] "At operation 206, the generative AI model generates a result using the prompt as an input.")
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (see Padgett paragraph [0064] "The similarity-assessment layer 104 may compare the output to each item in the repository of pre-existing content 106 in turn. With each comparison the layer 104 may calculate a similarity measure between the output and the item.", and see Padgett paragraph [0077] "In some implementations, both the result and the item from the repository are processed through a simplifying filter and the filtered results are then compared to assess the distance metric.", and see Padgett paragraph [0113] "In some cases, the repository 406 contains the original training data set, such that the filtering operation excludes outputs that too closely resemble the training data, e.g. instances of rote learning or close-to rote learning.")
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and (see Padgett paragraph [0083] "In this example the similarity measure is compared to a threshold value in operation 210. If the index exceeds the threshold, then it is deemed "too similar" to one of the items from the repository. If below the threshold, then it is sufficiently different.")
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and (see Padgett paragraph [0083] "However similarity is assessed, if the result is not "too similar", then it is output at operation 214. The output of the result may include transmitting the result to a user device for display and/or storage on the user device in some cases.")
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (see Padgett paragraph [0084] "If, however, the result is determined to be "too similar" based on the similarity measure in operation 210, then the system adjusts the input to the AI model in operation 212 and returns to re-run the model so as to generate and evaluate new results.")
Padgett does not explicitly teach "obtaining a summarization data package for the inference using at least a summarization model"
However, Kryscinski teaches
obtaining a summarization data package for the inference using at least a summarization model; (see Kryscinski paragraph [0016] "Embodiments described herein provide a two-step QFS model, which includes an extractor model to extract parts of the source document relevant to the input query, and an abstractor model to synthesize the extracted segments into a final summary.", and see Kryscinski paragraph [0045] "At step 418, an abstractor model (e.g., 180 in FIG. 1B) may synthesize the extracted at least one part into a final summary of the testing source document.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett to include obtaining a summarization data package for the inference using at least a summarization model, as taught by Kryscinski, in order to produce condensed summaries focused on the information of interest, thereby enabling faster and more focused comparison of lengthy generated outputs against the repository of pre-existing content, with predictable results and a reasonable expectation of success, see Kryscinski paragraph [0015].
As to claim 18, Padgett as modified by Kryscinski teaches the non-transitory machine-readable medium of claim 17, wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object. (see Kryscinski paragraph [0025] "FIG. 1B shows an example structure of the two-step QFS model that consists of an extractor model 160, which extracts parts of the source document 112 relevant to the input query 114, and an abstractor model 180, which synthesizes the extracted segments into a final summary 185 … In one embodiment, the abstractor model 180 may be a BART-large model.", and see Kryscinski paragraph [0053] "Model weights were initialized from pre-trained checkpoints available through the Huggingface Model Hub.")
As to claim 19, Padgett teaches a data processing system, comprising: a processor; and a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing a generative inference model, the operations comprising: (see Padgett paragraph [0131] "The above-described methods may be implemented by way of a suitably programmed computing device. FIG. 6A is a high-level operation diagram of an example computing device 605. The example computing device 605 may include a processor 600, a memory 610, an input interface 620, an output interface 630, and a communications subsystem 640.", and see Padgett paragraph [0132] "The processor 600 is a hardware processor. The processor 600 may, for example, be one or more ARM, Intel x86, PowerPC processors or the like.", and see Padgett paragraph [0137] "Software comprising instructions is executed by the processor 600 from a computer-readable medium. For example, software may be loaded into random-access memory from persistent storage of memory 610.")
obtaining an inference generated using the generative inference model and ingest data; (see Padgett paragraph [0062] "The generative AI model 102 is configured to take an input prompt, typically in text form but may also possibly include images or other media inputs. The model 102 creates an output related to the input prompt.", and see Padgett paragraph [0076] "At operation 206, the generative AI model generates a result using the prompt as an input.")
obtaining, using at least in part the summarization data package for the inference, levels of similarity of the inference with respect to reference works; (see Padgett paragraph [0064] "The similarity-assessment layer 104 may compare the output to each item in the repository of pre-existing content 106 in turn. With each comparison the layer 104 may calculate a similarity measure between the output and the item.", and see Padgett paragraph [0077] "In some implementations, both the result and the item from the repository are processed through a simplifying filter and the filtered results are then compared to assess the distance metric.", and see Padgett paragraph [0113] "In some cases, the repository 406 contains the original training data set, such that the filtering operation excludes outputs that too closely resemble the training data, e.g. instances of rote learning or close-to rote learning.")
making a determination regarding whether the inference is acceptable based on the levels of similarity and a similarity threshold; and (see Padgett paragraph [0083] "In this example the similarity measure is compared to a threshold value in operation 210. If the index exceeds the threshold, then it is deemed "too similar" to one of the items from the repository. If below the threshold, then it is sufficiently different.")
in a first instance of the determination where the inference is acceptable: providing the inference to a downstream consumer as a computer-implemented service; and (see Padgett paragraph [0083] "However similarity is assessed, if the result is not "too similar", then it is output at operation 214. The output of the result may include transmitting the result to a user device for display and/or storage on the user device in some cases.")
in a second instance of the determination where the inference is unacceptable: initiating performance of an action set to manage an impact of similarities between the inference and at least one of the reference works. (see Padgett paragraph [0084] "If, however, the result is determined to be "too similar" based on the similarity measure in operation 210, then the system adjusts the input to the AI model in operation 212 and returns to re-run the model so as to generate and evaluate new results.")
Padgett does not explicitly teach "obtaining a summarization data package for the inference using at least a summarization model"
However, Kryscinski teaches
obtaining a summarization data package for the inference using at least a summarization model; (see Kryscinski paragraph [0016] "Embodiments described herein provide a two-step QFS model, which includes an extractor model to extract parts of the source document relevant to the input query, and an abstractor model to synthesize the extracted segments into a final summary.", and see Kryscinski paragraph [0045] "At step 418, an abstractor model (e.g., 180 in FIG. 1B) may synthesize the extracted at least one part into a final summary of the testing source document.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett to include obtaining a summarization data package for the inference using at least a summarization model, as taught by Kryscinski, in order to produce condensed summaries focused on the information of interest, thereby enabling faster and more focused comparison of lengthy generated outputs against the repository of pre-existing content, with predictable results and a reasonable expectation of success, see Kryscinski paragraph [0015].
As to claim 20, Padgett as modified by Kryscinski teaches the data processing system of claim 19, wherein the summarization model comprises a foundation model adapted to extract a subset of information from a source data object. (see Kryscinski paragraph [0025] "FIG. 1B shows an example structure of the two-step QFS model that consists of an extractor model 160, which extracts parts of the source document 112 relevant to the input query 114, and an abstractor model 180, which synthesizes the extracted segments into a final summary 185 … In one embodiment, the abstractor model 180 may be a BART-large model.", and see Kryscinski paragraph [0053] "Model weights were initialized from pre-trained checkpoints available through the Huggingface Model Hub.")
Claim(s) 7, and 10-13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Padgett et al. (US 20240160902 A1) in view of Kryscinski et al. (US 20220277135 A1) and Franco-Salvador et al. (A systematic study of knowledge graph analysis for cross-language plagiarism detection, 15 January 2016).
As to claim 7, Padgett as modified by Kryscinski teaches the method of claim 6, further comprising:
Padgett does not explicitly teach "obtaining a structured representation for the inference based on the summarization data package for the inference using a structured representation schema", and "wherein the structured representation for the inference is used during the obtaining of the of the levels of similarity as a basis of comparison for the inference to the reference works"
However, Franco-Salvador teaches
obtaining a structured representation for the inference based on the summarization data package for the inference using a structured representation schema, (see Franco-Salvador section [3] "A knowledge graph is a weighted and directed graph that expands and relates the concepts belonging to a text.", and see Franco-Salvador section [3.2] "Initially we process a text fragment d with tokenization, multi-word extraction, part-of-speech (POS) tagging, and lemmatization to obtain the list of tuples (lemma,tag) T.", and see Franco-Salvador section [3.2] "We populate the vertex set V with the set SK of all the synsets in BabelNet which contain any <lemma,tag> tuple in T in the text fragment language L")
wherein the structured representation for the inference is used during the obtaining of the of the levels of similarity as a basis of comparison for the inference to the reference works. (see Franco-Salvador section [4] "Given a source document dL in a language L and a suspicious document d′L′ in a language L′, we compare documents in a four-step process", and see Franco-Salvador section [4] "For each pair of graphs (G, G′), G ∈ GC and G′ ∈ GC′, we adapt the algorithm of Montes y Gómez et al. (2001) to compare their similarity and to obtain the set of similarities SG between graph pairs.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett as modified by Kryscinski to include obtaining a structured representation for the inference using a structured representation schema and using it as the basis of comparison to the reference works, as taught by Franco-Salvador, in order to compare texts at the level of their concepts and the relationships between them rather than their exact wording, thereby enabling the similarity assessment to detect reuse even when the generated output has been reworded or paraphrased relative to the pre-existing content, with predictable results and a reasonable expectation of success, see Franco-Salvador section [3.4.2].
As to claim 10, Padgett as modified by Kryscinski teaches the method of claim 1, further comprising:
prior to obtaining the levels of similarity: for a portion of the reference works: obtaining a summarization data package for the portion of the reference works using at least the summarization model; and (see Kryscinski paragraph [0025] "FIG. 1B shows an example structure of the two-step QFS model that consists of an extractor model 160, which extracts parts of the source document 112 relevant to the input query 114, and an abstractor model 180, which synthesizes the extracted segments into a final summary 185.")
Padgett-Kryscinski does not explicitly teach "obtaining a structured representation for the portion of the reference works based on the summarization data package for the portion of the reference works using a structured representation schema."
However Franco-Salvador teaches
obtaining a structured representation for the portion of the reference works based on the summarization data package for the portion of the reference works using a structured representation schema. (see Franco-Salvador section [4] "In order to detect plagiarised sections of text between the documents dL and d′L′, we first segment them to obtain the sets of fragments FL and F′L′. We use a 5-sentence sliding window with a 2-sentence step to make the segmentation into fragments.", and see Franco-Salvador section [4] "We next use the method described in Section 3.2 to create the graph collections GC and GC′ of the text fragments FL and F′L′.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett as modified by Kryscinski to include obtaining, prior to the similarity determination, structured representations for portions of the reference works based on their summarization data packages using a structured representation schema, as taught by Franco-Salvador, in order to prepare fragment-level representations of the repository content in a preprocessing stage, thereby enabling fast portion-by-portion comparison at screening time and localization of which portions of the reference works the generated output resembles, with predictable results and a reasonable expectation of success, see Franco-Salvador section [5.5.2].
As to claim 11, Padgett-Kryscinski as modified by Franco-Salvador teaches the method of claim 7, wherein the structured representation for the inference comprises a graph-structured data model that specifies relationships between elements of the summarization data package for the inference, the graph-structured data model comprising nodes and edges, and the edges being based on the relationships between the elements associated with the edges. (see Franco-Salvador section [3] "A knowledge graph is a weighted and directed graph that expands and relates the concepts belonging to a text.", and see Franco-Salvador section [3.2] "That is, we add all the path vertices and edges to G.", and see Franco-Salvador section [3.3] "Relations in BabelNet are weighted to quantify the strength of the association between synsets. Knowledge graphs use these weights in order to weight their relations.")
As to claim 12, Padgett-Kryscinski as modified by Franco-Salvador teaches the method of claim 11, wherein obtaining the levels of similarity comprises:
performing a sub-graph analysis of the structured representation for the inference with respect to portions of structured representations for the reference works to identify whether a portion of the structured representation for the inference substantially matches one of the portions of the structured representations for the reference works. (see Franco-Salvador section [4] "For each pair of graphs (G, G′), G ∈ GC and G′ ∈ GC′, we adapt the algorithm of Montes y Gómez et al. (2001) to compare their similarity and to obtain the set of similarities SG between graph pairs. We calculate the similarity between the concepts in the two graphs using Dice's coefficient", and see Franco-Salvador section [4] "Basically, for each text fragment of dL we obtain PG, i.e., the top 5 most similar fragments of document d′L′ (line 3).", and see Franco-Salvador section 4 "Finally, we select as plagiarism the cases which combine more than thres2 text fragments (line 9).")
As to claim 13, Padgett as modified by Kryscinski teaches the method of claim 1,
Padgett-Kryscinski does not explicitly teach "wherein the levels of similarity indicate likelihoods that the inference plagiarizes the reference works."
However, Franco-Salvador teaches
wherein the levels of similarity indicate likelihoods that the inference plagiarizes the reference works. (see Franco-Salvador section [3.4.2] "If the text has been modified, it is quite likely having an intersection between the expanded concepts of the original text and the plagiarised one.", and see Franco-Salvador section [4] "Once we obtain the set SG with the similarities between the text fragments of the documents dL and d′L′, we employ the method introduced in Barrón-Cedeño (2012) and Barrón-Cedeño et al. (2013) to analyse the values and determine which fragments of text are cases of plagiarism.")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett as modified by Kryscinski such that the levels of similarity indicate likelihoods that the inference plagiarizes the reference works, as taught by Franco-Salvador, in order to analyze the computed similarity values to determine which portions of text constitute cases of improper reuse of the pre-existing works, thereby directly identifying the outputs that pose the reuse risk Padgett's screening is intended to avoid rather than reporting generic similarity alone, with predictable results and a reasonable expectation of success, see Franco-Salvador section [4].
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Padgett et al. (US 20240160902 A1) in view of Kryscinski et al. (US 20220277135 A1) and Franco-Salvador et al. (A systematic study of knowledge graph analysis for cross-language plagiarism detection, 15 January 2016) and further in view of Larson et al. (US 20250131289 A1).
As to claim 8, Padgett-Kryscinski as modified by Franco-Salvador teaches the method of claim 7,
Padgett does not explicitly teach "wherein the structured representation schema is adapted to facilitate identification of at least one of: a location indicated by the inference; a character indicated by the inference; an object indicated by the inference; and a law of nature indicated by the inference."
However, Larson teaches
wherein the structured representation schema is adapted to facilitate identification of at least one of: a location indicated by the inference; a character indicated by the inference; an object indicated by the inference; and a law of nature indicated by the inference. (see Larson paragraph [0021] "Thus, in the illustrated example of metadata extraction 208, the process identifies entities in the dataset 102 as POK, Sylvia Marek, Lucio Jakeb, and Save Our Wildlands.", and see Larson paragraph [0055] "Some of the goals with graph traversal are to find significant background context about relevant entities (e.g., events, people, places, etc.)")
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Padgett-Kryscinski as modified by Franco-Salvador to include a structured representation schema adapted to facilitate identification of entities such as people and places, as taught by Larson, in order to capture significant context about the relevant entities appearing in the text within the graph representation, thereby producing more informative structured representations whose comparison better reflects reuse of the same characters and locations across texts, with predictable results and a reasonable expectation of success, see Larson paragraph [0055].
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABDULLAH K ABOUD whose telephone number is (571)272-0025. The examiner can normally be reached Mon-Fri 8am-5pm.
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, Li B Zhen, can be reached at (571) 272-3768. 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.
/ABDULLAH KHALED ABOUD/ Examiner, Art Unit 2121
/Li B. Zhen/ Supervisory Patent Examiner, Art Unit 2121
Prosecution Insights