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 .
Claim Rejections - 35 USC § 101
Applicant’s amendments and arguments filed 3/12/26 suffice to obviate the 35 U.S.C. 101 rejection of the claims because the claimed domain specific training or adaptation of first and second low-rank matrices based on first and second loss functions associated with first and second domain specific tasks to improve a second model and thus the generated textual output thereof is considered to integrate the claimed elements into a practical application.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-20 rejected under 35 U.S.C. 103 as being unpatentable over Pena Pena: 11861884 hereinafter Pen further in view of Kalluri: 20210141861 hereinafter Kal and further in view of Sandler: 20200104706 hereinafter San and further in view of Hu: LORA: Low Rank Adaptation of Large Language Models (copy provided by Examiner, copyright 10/2021 and hereinafter Hu).
Regarding claim 1
Pen teaches:
A method comprising:
using a first machine learning model, generating a pseudo label associated with domain- specific training data (Pen: Col 2:61-2:64, 3:28-3:45, 6:24-6:30; Fig 2: pseudo labels generated in association with domain specific entity training data to thereby adapt characteristics of a labelled domain upon an unlabeled target domain, in this case the second machine learning model 210 functions in the manner claimed with respect to the first model, such as for text information extraction) comprising a pseudo label associated with a first domain- specific task and a pseudo label associated with a second domain-specific task (Pen: Abstract; Col 2:28-2:42: system performs key information extraction (KIE) for generation of the pseudo labels based on information classified upon a specific domain, tasks thereof such as utilizing classification types to reify the domain; e.g. the determination of domain tasks relevant to determined KIE such as goods purchased by vendor upon an accounting or tax reporting domain),
wherein the pseudo label comprises machine-generated text extracted from the domain-specific training data (Pen: Col 5:14-5:20, 9:44-9:52; Fig 1, 2: such as by utilizing prototype labels extracted from a receipt such as depicted in figure 1, such as a name, an address, an item type, etc.); and
fine-tuning a second machine learning model, to generate text using a domain-specific document (Pen: Col 3:45-3:60, 5:32-5:42, etc.; Fig 2: a model subsequent to and generated from the first model, domain specific data, pseudo labels, etc. thereof used to improve the subsequent model for adaptability on a target or subsequent domain, such as by using documents upon the domain, prototype labelling thereof) wherein the fine-tuned second machine learning model generates text of the content type from a domain-specific data (Pen: Col 1:40-1:45, 7:59-8:5, 8:19-25, 8:31-8:34; Fig 2, 3: such as by iteratively improving the textual content of pseudo labels of entities, characteristics thereof and thereby determining additional documents to which to apply pseudo labels).
Pen does not explicitly teach adapting the classification, pseudo label, etc. pipeline by the use of first and second low rank matrices to accomplish the claimed fine-tuning a second machine learning model using the pseudo label, the domain-specific training data, a first low-rank weight matrix, and a second low-rank weight matrix. Additionally, Pen teaches or suggests generation of text of the content type by reification of the textual content of pseudo labels but does not discuss generating new textual labels and/or new text based on the first domain nor does Pen address steps by which fine- tuning the second machine learning model comprises: updating a first low-rank weight matrix and a second low-rank weight matrix of the second machine learning model associated with a first domain-specific task using a first loss function and the pseudo label associated with the first domain-specific task, and updating a first low-rank weight matrix and a second low-rank weight matrix of the second machine learning model associated with a second domain- specific task using a second loss function and the pseudo label associated with the second domain-specific task.
In a related field of endeavor Kal teaches a system, method, etc. for adapting generalized vocabulary items in a trained model to estimate unknown labels for new data examples (Kal: ¶ 30, 31, 97, 102, etc.; Fig 5, 6: “document may be transformed into an ML feature vector based at least in part on occurrence of vocabulary tokens;” to thereby “estimate unknown labels for new examples based on patterns in feature vectors and the associated labels”) comprising fine tuning a model trained upon a particular vocabulary using a different vocabulary (Kal: ¶ 36; Fig 3, 5: this is considered to the domain abstraction as recited and similarly taught in Pen supra) wherein features of tokens are vectorized, weighted, etc. and additionally used for training (Kal: 35, 36, 81-83) thereby operative for determining plural content types, features, etc. such as for adapting to plural vocabularies, unknown labels, thereof, etc. thereby assigning labels to same (Kal: ¶ 81-83, 97; 3, 5: such as exemplified by claim 12: “receiving a new document with an unknown label; generating a new feature vector for the new document as a function of the fixed-length vocabulary; and estimating, by the trained machine learning model, the unknown label based at least in part on the new feature vector,”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to preprocess the unlabeled data of Pen in concert with the structured vocabulary based tokenized features of Kal to generate new textual pseudo labels to thereby adapt the pseudo labels of Pen upon new domains, specific vocabularies, etc. as taught or suggested by Kal and for at least the purpose of tokenizing, weighting and/or vectorizing textual pseudo labels of unlabeled data for generating additional, improved, etc. pseudo labels encompassing separate domains, vocabularies, etc.; one of ordinary skill in the art would have expected only predictable results therefrom.
Pen in view of Kal does not explicitly teach adapting the classification, pseudo label, etc. pipeline by fine- tuning the second machine learning model comprises: updating a first low-rank weight matrix and a second low-rank weight matrix of the second machine learning model associated with a first domain-specific task using a first loss function and the pseudo label associated with the first domain-specific task, and updating a first low-rank weight matrix and a second low-rank weight matrix of the second machine learning model associated with a second domain- specific task using a second loss function and the pseudo label associated with the second domain-specific task.
In a related field of endeavor San teaches a system and method for fine tuning a model in the form of modifying a task specific model to include a model patch comprising a second set of learnable parameters for a second task (San: Abstract; ¶ 6, 58) wherein a first, second, etc. model comprises task specific data and wherein the addition of the patch adjusts, fine tunes, etc. a first, second, etc. model to adapt to a second task based on, with respect to, etc. the model patch such as with respect to minimizing loss thereof (San: Abstract; ¶ 6, 22-25, 58, 70, 85; Fig 1A: a first, second model adjusted with respect to patch such as by minimizing loss over a series of gradient descent iterations). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to modify the Pen in view of Kal system and method to fine tune a model to particular domain specific tasks without fully retraining the entire model for at least the purpose of adapting a document model to a second domain specific task by patching and updating the model based on calculating loss with respect thereto as taught or suggested by San; one of ordinary skill in the art would have expected only predictable results therefrom.
Pen in view of Kal in view of San teach adapting the classification, pseudo label, etc. pipeline by the use of first and second low rank matrices to accomplish the claimed fine-tuning a second machine learning model using the pseudo label, the domain-specific training data, a first low-rank weight matrix, and a second low-rank weight matrix.
In a related field of endeavor Hu teaches a system and method for adapting a pretrained model using a first and second low rank matrices such as for the purpose of the recited fine tuning for improving a model by iteratively updating a first and second low rank weight matrix such as with respect to a loss function(s) (Hu: Abstract; Introduction; Fig 1: A and B; Table 18: matrices comprise first and second low rank matrices for large language model adaptation validated based on an iterated loss functions). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to combine the Pen in view of Kal in view of San system, method, etc. and the teachings of Hu to thereby tune or fine tune a model, subsequent model, etc. using a multi stage training pipeline comprising determining pseudo labels such as with respect to a confidence score and confidence calibration, thresholding, etc. as taught or suggested by Pen in view of Kal in view of San and to adapt the loss based fine tuning thereof by adapting the pipeline to utilize the low rank matrices taught or suggested by Hu to thereby train a reduced set of parameters as taught or suggested by Hu with respect to a reified subset of pseudo labelled data with improved calibration based on the application of specific loss calculations for the first and second Hu matrices as taught or suggested by Pen in view of Kal in view of San for at least the purpose of reducing computational costs of the Pen in view of Kal in view of San taught fine tuning by compacting the data sizes using low rank matrices to generate domain specific outputs using fewer parameters, smaller structures, etc.; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 2
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 1, further comprising: generating, by the first machine learning model, a first training data set of a first size, wherein the first training data set comprises a plurality of pseudo labels paired with domain-specific documents (Pen: Col 1:27-1:40, 3:14-3:18, 4:6-4:17: a small strongly labelled domain dataset comprising documents and entity labeling of textual instances therein is augmented by generated new input label pairs); (Kal: ¶ 34-36, 86, 87, 97, etc.; Fig 3, 5, 6: labeled training data comprises documents and labels applied thereto to adapt a model tuned on a first vocabulary of a set of training to an additional second vocabulary in an additional document instance); and fine-tuning the second machine learning model using the first training data set of the first size (Pen: Col 3:14-3:18, 3:45-4:35; Fig 2: system operates to fine tune based on a set of strongly labelled data, additional pseudo labelled data, etc.); (Kal: ¶ 33-36, 81-83, 97, Claim 12: in this way the system tunes the labelling to adapt the first vocabulary to the second vocabulary by augmenting the set of tokens from a first set to a second set based on additional document instances). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 3
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 2, wherein the second machine learning model is pretrained on a second training data set of a second size, and the first size of the first training data set is less than the second size of the second training data set (Pen: Col 3:14-3:18, 3:45-4:35; Fig 2: system operates to fine tune based on a set of strongly labelled data, additional pseudo labelled data, etc.); (Kal: ¶ 33-36, 81-83, 97, Claim 12; Fig 3, 5, 6: the system tunes the labelling to adapt the first vocabulary to the second vocabulary by augmenting the set of tokens from a first set to a second set based on additional document instances wherein the second set comprises a size augmented set of tokens). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 4
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 1, further comprising: accessing the second machine learning model pretrained on domain-neutral data (Pen: 5:45-5:50: a general model pretrained on general data for entity recognition), (Kal: Title; Abstract: generalized vocabulary tokens, model based thereon determines second, additional subsequent, etc. model fine tuned to include additional vocabulary tokens from an additional dataset of documents, labels, etc.); (Hu: Abstract: system adapts a first, general domain data model, llm, etc. to a second set of tasks or domains) wherein the second machine learning model comprises a plurality of pretrained weights in a pretrained weight matrix (Hu: § 4: such as a frozen matrix of weights adaptable to train, fine tune, etc. a second, subsequent, etc. model based on an adaptive relationship between first and second low rank matrices). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 5
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 1, wherein the first machine learning model comprises a large language model and wherein the second machine learning model comprises a large language model (Pen: 5:45-5:50; Fig 2: a general model pretrained on general data for entity recognition and adapted to create a subsequent fine-tuned domain model); (Kal: Title; Abstract: generalized vocabulary tokens, model based thereon determines second, additional subsequent, etc. model fine tuned to include additional vocabulary tokens from an additional dataset of documents, labels, etc.); (Hu: Abstract: system adapts a first, general domain data model, llm, etc. to a second set of tasks or domains). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 6
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 1, wherein the domain-specific training document is an unstructured document. Examiner considers the claimed subject matter obvious to try. It is well known that datasets, documents may be structured or unstructured and the cited art generally orbits the problem of adapting a domain model trained on a dataset of labelled and pseudo labeled documents upon a second domain of distinct and unlabeled documents. There exist a plurality of means for accomplishing such a task including using structured documents, using unstructured documents and/or a combination thereof these means comprise a finite number of predictable potential solutions which one of ordinary skill in the art could have pursued with a reasonable expectation of success. The claim is thus considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 7
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 1, wherein fine-tuning the second machine learning model comprises defining the first low-rank weight matrix and the second low-rank weight matrix, and the method further comprises: storing the defined first low-rank weight matrix and the defined second low-rank weight matrix associated with the first domain-specific task; storing the first low-rank weight matrix and the second low-rank weight matrix associated with the second domain-specific task. (San: Abstract; ¶ 6, 22-25, 58, 70, 85; Fig 1A: system iteratively updates first, second, etc. patched models based on loss determination which requires data structures in computed and stored). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claim 8
Pen in view of Kal in view of San in view of Hu teaches or suggests:
The method of claim 7, further comprising: inputting, to the second machine learning model comprising a plurality of pretrained weights in a pretrained weight matrix, a document to obtain text of the content type from the document, wherein the second machine learning model further comprises an adaptation component comprising the defined first low-rank weight matrix and the defined second low-rank weight matrix (Pen: Col 1:27-1:40, 3:14-3:18, 3:45-4:35, 9:35-9:49; Fig 2: such as the model to be fine-tuned which receives additional documents to adaptively label); (Kal: ¶ 33-36, 81-83, 97, 102,: fig 3, 5, 6: such providing to the model additional documents comprising a vocabulary, set of labels, etc. to be adapted upon the model); (Hu: § 4: such as the frozen weights of the W matrix using the first, second low rank matrices). The claim is considered obvious over Pen as modified by Kal, San, and Hu as addressed in the base claim as it would have been obvious to apply the further teaching of Pen, Kal, San, and/or Hu to the modified device of Pen, Kal, San, and Hu; one of ordinary skill in the art would have expected only predictable results therefrom.
Regarding claims 9, 16—the claims are considered to recite substantially similar subject matter to that of claim 1 and are similarly rejected.
Regarding claims 10, 17—the claims are considered to recite substantially similar subject matter to that of claim 2 and are similarly rejected.
Regarding claims 11, 18—the claims are considered to recite substantially similar subject matter to that of claim 3 and are similarly rejected.
Regarding claims 12, 20—the claims are considered to recite substantially similar subject matter to that of claim 5 and are similarly rejected.
Regarding claim 13—the claim is considered to recite substantially similar subject matter to that of claim 6 and is similarly rejected.
Regarding claim 14—the claim is considered to recite substantially similar subject matter to that of claim 7 and is similarly rejected.
Regarding claims 15, 19—the claims are considered to recite substantially similar subject matter to that of claim 8 and are similarly rejected.
Response to Arguments
Applicant’s arguments and claim amendments, see Remarks and Claims, filed 3/12/26, with respect to the rejection(s) of claim(s) 1-20 under 345 USC 103 over Pena, Kalluri, and Hu have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Pena, Kalluri, Sandler, and Hu.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PAUL C MCCORD whose telephone number is (571)270-3701. The examiner can normally be reached 730-630 M-F.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, CAROLYN EDWARDS can be reached at (571) 270-7136. 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.
/PAUL C MCCORD/Primary Examiner, Art Unit 2692
Prosecution Insights