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 .
Response to Arguments
Applicant’s arguments filed 4 August 2025 have been fully considered.
Section 101
Applicant describes its invention as handling a meta data extraction process “in the same analogy as the human brain works.” Specification ¶ [002]. It is unclear how Applicant possibly hopes to prevail on an argument that the disclosed invention is not mere automation of a mental process.
Applicant argues that capturing a deep contextual meaning from one or more text cells present in the document using a domain specific language model is not practically performable in the human mind. Examiner respectfully disagrees. Under a broadest reasonable interpretation, the “domain specific language model” is a generic machine learning function. A claim limitation reciting using a generic machine learning function to perform a mental process recites the mental process. MPEP § 2106.04(a)(2)(III)(C); see Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205, 1208 (Fed. Cir. 2025).
Applicant argues that capturing a complex visual layout of the document using a domain specific visual model is not practically performable in the human mind. Examiner respectfully disagrees. Under a broadest reasonable interpretation, the “domain specific visual model” is a generic machine learning function. A claim limitation reciting using a generic machine learning function to perform a mental process recites the mental process. MPEP § 2106.04(a)(2)(III)(C); see Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205, 1208 (Fed. Cir. 2025).
Applicant argues that performing substitution of missing sub-words with already available sub-words in a dictionary is not practically performable in the human mind. Examiner respectfully disagrees. A person can think if they recognize a word they know.
Applicant argues that passing a masked token together with context to a MLM is not practically performable in the human mind. Examiner agrees. It is, however, mere instruction to apply the mental process of “using context to guess a missing word” using a generic language model as a tool.
Applicant argues that calculating m-gram minimum edit distance is not a calculation. Examiner respectfully disagrees. Firstly, the grounds of rejection with respect to this limitation is not that the invention is a calculation but rather that this limitation is. To argue that it is integrated into a practical application is not an argument that it is not a calculation.
Applicant argues that processing captured deep contextual meaning and complex visual layout to determine meta-data information is not practically performable in the human mind. Examiner respectfully disagrees. Applicant describes this as practically performable in the human mind. Specification [002] (“It is possible to make this meta data extraction process more advance if we may handle it in the same analogy as the human brain works. For example, it looks on all the aspects of the information present in the document, such as the information may be available in terms of size, font, position, visual, and context etc.”). Using generic “deep learning” computer functions to automate this mental process is not patentable. MPEP §§ 2106.05(a), 2106.05(f).
Applicant argues that the invention integrates the recited abstract idea in a practical application. Examiner respectfully disagrees. The invention is mere instruction to perform the recited abstract idea on a computer using generic computing functionality, i.e., using a computer as a tool to perform the recited abstract idea. MPEP § 2106.05(f). Detecting linguistic and visual contexts is abstract. Generating metadata information that reflects both the content and structure of the document, supporting advanced tasks such as classification, information retrieval, and workflow automation is abstract. Determining a style embedding is abstract. Pre-processing, segmentation, object detection, coordinate extraction, and contextual embedding are all abstract. No implementation details are disclosed, much less recited.
The “advanced text tokenization” fails to improve the technology of OCR because it is equally applicable to other areas, such as fixing typos in word processing, such that it is clear that any purported improvement lies in the abstract idea itself and its inclusion here generally links the abstract idea to performance in the technological field of OCR.
Applicant argues that the claim recites significantly more than the abstract idea. Examiner respectfully disagrees. Using generic machine learning to perform an abstract idea is claiming the abstract idea itself. See Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205, 1215 (Fed. Cir. 2025).
Section 103
Applicant argues that Li does not teach capturing, by the meta-data extraction device, a deep contextual meaning from the one or more text cells present in the document using a domain specific language model or capturing, by the meta-data extraction device, a complex visual layout of the document using a domain specific visual model. Examiner respectfully disagrees. As Applicant admits, “Li employs an LSTM-based semantic feature extractor and CNN-derived visual features.” Whatever other differences there are between Applicant’s disclosure and the disclosure of Li, they exist in the disclosure, not the claims. Applicant is welcome to amend these features into the claims, but until this is done, Applicant may not rely on them to distinguish the invention as claimed under a broadest reasonable interpretation and Li.
Applicant argues that Liu does not teach calculating n-gram minimum edit distance for remaining OOV sub-words. Examiner agrees, but notes that the claims do not require that there actually be any OOV sub-words not substituted; the claims cover an embodiment where the advanced text tokenization performs token substitution on all the OOV tokens by finding all the missing sub-words in the vocabulary of a pre-trained model, in which case there are no OOV sub-words remaining to calculate an m-gram distance for.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-6, 8-14, and 16-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 9, and 17 recite the limitation “the vocabulary of pre-trained model.” There is insufficient antecedent basis for this limitation in the claim. For the purposes of examination, this limitation is read as “a vocabulary of a pre-trained model.”
Claims 1, 9, and 17 recite the limitation “the masked Language Model.” There is insufficient antecedent basis for this limitation in the claim. For the purposes of examination, this limitation is read as “a masked Language Model.”
Claims 1, 9, and 17 recite the limitation “the OOV sub-words not substituted by any in-vocabulary sub-word.” There is insufficient antecedent basis for this limitation in the claim. For the purposes of examination, this limitation is read as “for any OOV sub-words not substituted by any in-vocabulary sub-word,” with the understanding that this is an optional feature as the broadest reasonable interpretation of the disclosed invention does not require that there actually are any OOV sub-words not so substituted. Reading this instead as “for one or more OOV sub-words not substituted by any in-vocabulary sub-word” would impermissibly import unclaimed limitations from the specification.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-6, 8-14, and 16-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
As per claims 1, 9, and 17:
The claim(s) recites an abstract idea.
The limitation, “capturing, by a meta-data extraction device, one or more of style attributes from the document, where a style embedding is created by concatenating the captured one or more style attributes, and wherein the style embedding is created before creating a word embedding from text present in the document,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “capturing” encompasses a person forming a judgment as to what style attributes are used. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “identifying, by the meta-data extraction device, a cell-wise location coordinates for text characters associated with the document, wherein the cell-wise location coordinates are indicative of inter-relationship of one or more cells of the document based on surrounding embedding,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “identifying” encompasses a person forming a judgment as to position of text with relation to other text on a page. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “capturing, by the meta-data extraction device, a deep contextual meaning from the one or more text cells present in the document using a domain specific language model,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “capturing” encompasses a person forming a judgment as to the semantic meaning of the texts. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “capturing, by the meta-data extraction device, a complex visual layout of the document using a domain specific visual model,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “capturing” encompasses a person forming a judgment as to what meaning can be gleaned from the visual layout. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “processing, by the meta-data extraction device, the captured deep contextual meaning and the complex visual layout to determine the meta-data information,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “processing” encompasses a person forming a judgment as to the content of the meta-data information. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “substituting a missing sub-word with already available sub-word in the vocabulary of pre-trained model,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “substituting” encompasses a person forming a judgment as to what available sub-word to use. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “substituting OOV sub-words by mask tokens,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “substituting” encompasses a person forming a judgment as to what mask token to use. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
The limitation, “for the OOV sub-words not substituted by any in-vocabulary sub-word, calculating m-gram minimum edit distance between remaining OOV sub-words and in-vocabulary sub-words to determine a substitution word for OOV sub-words,” as drafted, covers a calculation. This limitation therefore falls within the “Mathematical Concepts” grouping of abstract ideas. MPEP § 2106.04(a)(2)(I).
Accordingly, the claim(s) recites abstract ideas. MPEP § 2106.04(a).
For the purposes of evaluating whether the claim(s) is directed to an abstract idea or is significantly more than an abstract idea, these recited abstract ideas can be considered together as a single abstract idea, namely extracting text and metadata from a document. MPEP § 2106.04(II)(B).
The additional element, “passing the mask tokens to the masked Language Model along with its context,” merely invokes machine learning as a tool to perform the mental process of figuring out what token to substitute for the out-of-vocabulary token. MPEP § 2106.05(f).
As an ordered combination, the claim(s) merely automates the manual task of determining a replacement token using generic computer functions. MPEP §§ 2106.05(a), 2106.05(f).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claim(s) as a whole, and therefore the claim is directed to an abstract idea. MPEP § 2106.04(d).
As discussed above with respect to integration of the abstract idea into a practical application, the conclusions for the additional elements being generic computer components and mere instructions to apply on a computer, insignificant extra-solution activity, and/or mere field of use limitations are carried over and these additional elements do not provide significantly more than the abstract idea. MPEP § 2106.05(II).
As an ordered combination, the claim simply appends well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the abstract idea because the claim as a whole amounts to nothing more than generic computer functions merely used to implement the abstract idea. MPEP §§ 2106.07(a)(III)(B), 2106.05(d)(II); see BASCOM Global Internet Servs. v. AT&T Mobility LLC, 827 F.3d 1341, 1349 (Fed. Cir. 2016).
Accordingly, the claim(s) does not recite additional elements, either individually or in combination, that amount to significantly more than the abstract idea. MPEP § 2106.05. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
As per claims 2, 10, and 18:
The claim(s) recites an abstract idea.
The limitation, “wherein the cell-wise location coordinates for the at least one attribute are captured using a page segmentation and a border table extraction mechanism,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “capturing” encompasses a person forming a judgment as to the location of page segments and border tables. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
Accordingly, the claim(s) recites an abstract idea. MPEP § 2106.04(a). As the claim(s) recites no additional elements, the abstract idea is not integrated into a practical application, the claim is directed to the abstract idea, and the claim(s) does not amount to significantly more than the abstract idea. MPEP § 2106.07. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
As per claims 3, 11, and 19:
The claim(s) recites an abstract idea.
The limitation, “wherein for determining the relationship between nearby one or more cells present in the document, a distance between each of the cells is determined, and wherein upon determination of the distance between each of the cells, a shortest distant text cell in top, left, right and bottom direction of the particular cell is determined,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “determining” encompasses a person forming a judgment as to the distances. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
Accordingly, the claim(s) recites an abstract idea. MPEP § 2106.04(a). As the claim(s) recites no additional elements, the abstract idea is not integrated into a practical application, the claim is directed to the abstract idea, and the claim(s) does not amount to significantly more than the abstract idea. MPEP § 2106.07. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
As per claims 4, 12, and 20:
The abstract idea of extracting information from a document is not integrated into a practical application.
The additional element, “wherein distance of each of top, left, right and bottom cell is determined with respect to a main text cell to create a compact surrounding embedding, wherein the surrounding embedding is created using a Graph Convolution Network with Informative Attention (GCN-IA),” amounts to no more than generally linking the abstract idea to the particular field of use or technological environment of machine learning. MPEP § 2106.05(h).
As an ordered combination, the claim(s) is mere instruction to apply the abstract idea on a computer. MPEP § 2106.05(f).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claim(s) as a whole, and therefore the claim is directed to an abstract idea. MPEP § 2106.04(d).
As discussed above with respect to integration of the abstract idea into a practical application, the conclusions for the additional elements being generic computer components and mere instructions to apply on a computer, insignificant extra-solution activity, and/or mere field of use limitations are carried over and these additional elements do not provide significantly more than the abstract idea. MPEP § 2106.05(II). In re-evaluating the limitations that are field of use, the following limitations represent elements that have been recognized as well-understood, routine, conventional activity within the field of computer functions:
The additional element, “wherein distance of each of top, left, right and bottom cell is determined with respect to a main text cell to create a compact surrounding embedding, wherein the surrounding embedding is created using a Graph Convolution Network with Informative Attention (GCN-IA),” is well-understood, routine, and conventional activity because it is described in a manner that indicates that the additional element is sufficiently well-known that the specification does not need to describe the particulars of such additional elements to satisfy 35 U.S.C. 112(a). MPEP § 2106.07(a)(III)(A).
As an ordered combination, the claim simply appends well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the abstract idea because the claim as a whole amounts to nothing more than generic computer functions merely used to implement the abstract idea. MPEP §§ 2106.07(a)(III)(B), 2106.05(d)(II); see BASCOM Global Internet Servs. v. AT&T Mobility LLC, 827 F.3d 1341, 1349 (Fed. Cir. 2016).
Accordingly, the claim(s) does not recite additional elements, either individually or in combination, that amount to significantly more than the abstract idea. MPEP § 2106.05. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
As per claims 5 and 13:
The claim(s) recites an abstract idea.
The limitation, “an erroneous output is corrected,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “correcting” encompasses a person forming a judgment as to the correct output. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
Accordingly, the claim(s) recites an abstract idea. MPEP § 2106.04(a).
The abstract idea is not integrated into a practical application.
The additional element, “wherein an erroneous output is corrected using an advanced post processing steps comprising at least one of a rule based post-processing mechanism, a hashed dictionary based rapid text alignment mechanism, a machine learning language model for alphabetic spelling correction, and a dimension unit conversion mechanism,” amounts to no more than generally linking the abstract idea to the particular field of use or technological environment of machine learning. MPEP § 2106.05(h).
As an ordered combination, the claim(s) is mere instruction to apply the abstract idea on a computer. MPEP § 2106.05(f).
Accordingly, the additional elements, individually or in combination, do not integrate the abstract idea into a practical application, even viewing the claim(s) as a whole, and therefore the claim is directed to an abstract idea. MPEP § 2106.04(d).
As discussed above with respect to integration of the abstract idea into a practical application, the conclusions for the additional elements being generic computer components and mere instructions to apply on a computer, insignificant extra-solution activity, and/or mere field of use limitations are carried over and these additional elements do not provide significantly more than the abstract idea. MPEP § 2106.05(II). In re-evaluating the limitations that are field of use, the following limitations represent elements that have been recognized as well-understood, routine, conventional activity within the field of computer functions:
The additional element, “wherein an erroneous output is corrected using an advanced post processing steps comprising at least one of a rule based post-processing mechanism, a hashed dictionary based rapid text alignment mechanism, a machine learning language model for alphabetic spelling correction, and a dimension unit conversion mechanism,” is well-understood, routine, and conventional activity because it is described in a manner that indicates that the additional element is sufficiently well-known that the specification does not need to describe the particulars of such additional elements to satisfy 35 U.S.C. 112(a). MPEP § 2106.07(a)(III)(A).
As an ordered combination, the claim simply appends well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the abstract idea because the claim as a whole amounts to nothing more than generic computer functions merely used to implement the abstract idea. MPEP §§ 2106.07(a)(III)(B), 2106.05(d)(II); see BASCOM Global Internet Servs. v. AT&T Mobility LLC, 827 F.3d 1341, 1349 (Fed. Cir. 2016).
Accordingly, the claim(s) does not recite additional elements, either individually or in combination, that amount to significantly more than the abstract idea. MPEP § 2106.05. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
As per claims 6 and 14:
The claim(s) recites an abstract idea.
The limitation, “wherein the extracted output represents linguistic and visual contexts of the document, and wherein the extracted output captures one or more of specific terms and design of the document using a detailed region information mechanism,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “searching” encompasses a person forming a judgment as to the semantic meaning of the secondary texts. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
As per claims 8 and 16:
The claim(s) recites an abstract idea.
The limitation, “wherein for reducing the out-of-vocabulary problem, one or more of secondary vocabulary is created for linking new words to existing words of the document during fine-tuning of an advanced language model for down streaming tasks,” as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. For example, in the context of this limitation, “creating” encompasses a person forming a judgment as to the content of the secondary vocabulary. This limitation therefore falls within the “Mental Processes” grouping of abstract ideas. MPEP § 2106.04(a)(2)(III).
Accordingly, the claim(s) recites an abstract idea. MPEP § 2106.04(a). As the claim(s) recites no additional elements, the abstract idea is not integrated into a practical application, the claim is directed to the abstract idea, and the claim(s) does not amount to significantly more than the abstract idea. MPEP § 2106.07. Therefore, as the claim(s) is directed to an abstract idea and does not recite additional elements that amount to significantly more than the abstract idea, the claim(s) is not patentable. MPEP § 2106.
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, 9-14, and 17-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li et al., GFTE: Graph-based Financial Table Extraction, in view of Anders et al., US 2022/0058336 A1, and Liu et al., WO 2022/169656 A1.
As per claims 1, 9, and 17, Li teaches:
identifying, by the meta-data extraction device, a cell-wise location coordinates for text characters associated with the document, wherein the cell-wise location coordinates are indicative of inter-relationship of one or more cells of the document based on surrounding embedding, Li pg. 5 (“the absolute locations”);
capturing, by the meta-data extraction device, a deep contextual meaning from the one or more text cells present in the document using a domain specific language model, Li pg. 5 (“the textual content”);
capturing, by the meta-data extraction device, a complex visual layout of the document using a domain specific visual model, Li pg. 5 (“the image”); and
processing, by the meta-data extraction device, the captured deep contextual meaning and the complex visual layout to determine the meta-data information, Li pg. 5, where the underlying tabular layout is metadata.
Li, however, does not teach:
capturing, by a meta-data extraction device, one or more of style attributes from the document, where a style embedding is created by concatenating the captured one or more style attributes, and wherein the style embedding is created before creating a word embedding from text present in the document.
The analogous and compatible art of Anders, however, teaches that a document can be classified also by font attributes. Anders ¶ 0041.
It would therefore have been obvious to one of ordinary skill in the art at the time of filing to determine style attributes as in Anders, where the representation of the embedding as a concatenation is a design choice (suggested by Li, pg. 5 (“We then concatenate”)), where the timing is a design choice, in order to better recognize the layout of a table within a PDF.
Neither Li nor Anders teach:
performing advanced text tokenization by identifying one or more out-of-vocabulary (OOV) tokens based on token substitution, wherein the token substitution comprises:
substituting a missing sub-word with already available sub-word in the vocabulary of pre-trained model;
substituting OOV sub-words by mask tokens;
passing the mask tokens to the masked Language Model along with its context; and
for the OOV sub-words not substituted by any in-vocabulary sub-word, calculating m-gram minimum edit distance between remaining OOV sub-words and in-vocabulary sub-words to determine a substitution word for OOV sub-words.
The analogous and compatible art of Liu, however, teaches substituting mask tokens in place of out of vocabulary tokens, then using a trained language model to substitute the unknown word. Liu ¶¶ 00039-42.
It would therefore have been obvious to one of ordinary skill in the art at the time of filing to modify the teachings of Li to use the advanced text tokenization by identifying one or more out-of-vocabulary (OOV) tokens based on token substitution of Liu in order to ensure that all words are recognized.
As per claims 2, 10, and 18, the rejection of claims 1, 9, and 17 is incorporated, and Li further teaches:
wherein the cell-wise location coordinates for the at least one attribute are captured using a page segmentation and a border table extraction mechanism, Li pg. 4 (“Cross-page table”).
As per claims 3, 11, and 19, the rejection of claims 1, 9, and 17 is incorporated, and Li further teaches:
wherein for determining the relationship between nearby one or more cells present in the document, a distance between each of the cells is determined, and wherein upon determination of the distance between each of the cells, a shortest distant text cell in top, left, right and bottom direction of the particular cell is determined, Li pg. 5 (“Meanwhile, it is not hard to notice that a table structure can be represented by far fewer edges, as long as a node is connected to its nearest neighbors including both vertical ones and horizontal ones. With the knowledge of node position, we are also capable of recovering the table structure from these relations.”).
As per claims 4, 12, and 20, the rejection of claims 1, 9, and 17 is incorporated, and Li further teaches:
wherein distance of each of top, left, right and bottom cell is determined with respect to a main text cell to create a compact surrounding embedding, wherein the surrounding embedding is created using a Graph Convolution Network with Informative Attention (GCN-IA), Li pg. 5 (“We concatenate the position feature and the text feature together and feed them to a two-layer graph convolutional network (GCN).”).
As per claims 5 and 13, the rejection of claims 1 and 9 is incorporated, but Li does not explicitly teach:
wherein an erroneous output is corrected using an advanced post processing steps comprising at least one of a rule based post-processing mechanism, a hashed dictionary based rapid text alignment mechanism, a machine learning language model for alphabetic spelling correction, and a dimension unit conversion mechanism.
Official notice is taken, however, that, at the very least, a machine learning language model for alphabetic spelling correction is known in the prior art.
It would therefore have been obvious to one of ordinary skill in the art to incorporate a machine learning language model for alphabetic spelling correction in interpreting a table in order to interpret it correctly.
As per claims 6 and 14, the rejection of claims 1 and 9 is incorporated, and Li further teaches:
wherein the extracted output represents linguistic and visual contexts of the document, and wherein the extracted output captures one or more of specific terms and design of the document using a detailed region information mechanism, Li pg. 5 (“three types of information are included, i.e. the textual content, the absolute locations and the image”).
As per claims 8 and 16, the rejection of claims 1 and 15 is incorporated, but Li does not teach:
wherein for reducing the out-of-vocabulary problem, one or more of secondary vocabulary is created for linking new words to existing words of the document during fine-tuning of an advanced language model for down streaming tasks.
The analogous and compatible art of Liu, however, teaches substituting mask tokens in place of out of vocabulary tokens, then using a trained language model to substitute the unknown word, where the process links new words to existing words in order to improve the knowledge encoder. Liu ¶¶ 00039-42.
It would therefore have been obvious to one of ordinary skill in the art at the time of filing to modify the teachings of Li to use the advanced text tokenization by identifying one or more out-of-vocabulary (OOV) tokens based on token substitution of Liu in order to ensure that all words are recognized.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM SPIELER whose telephone number is (571)270-3883. The examiner can normally be reached Monday-Friday, 11-3.
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, Ann Lo can be reached at 571-272-9767. 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.
WILLIAM SPIELER
Primary Examiner
Art Unit 2159
/WILLIAM SPIELER/ Primary Examiner, Art Unit 2159