DETAILED ACTION
This examination is in response to the communication filed on 12/15/2023. Claims 1-20 are currently pending, where claims 1, 13 and 20 are independent.
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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 12/15/2023 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Independent claims 1, 13, and 20 recite (a) “generating…a summary of a source document”; (b) “generating… a plurality of topics based on the summary and a predetermined number of topics”; (c) “generating… expanded text for each of the plurality of topics”; (d) “selecting an image from the source document for each of the plurality of topics by computing a similarity score between the image and the expanded text”; and (e) “generating a summary document including the plurality of topics, the expanded text, and the selected image.”
The limitations of “generating…”, “generating…”, “generating…”, “selecting…” and “generating…” as drafted, are a process that, under a broadest reasonable interpretation, covers the abstract idea of “mental processes” because they cover concepts performed in the human mind, including observation, evaluation, judgement and opinion. See MPEP 2106.04(a)(2). That is, other than reciting “at least one processor” (claim 13), “at least one memory” (claim 13), a “a language generation model” (claims 1, 13 and 20), nothing in the claimed elements preclude the steps from practically being performed by a person
generating
generating
generating
selecting an image from the source document for each of the plurality of topics by computing a similarity score between the image and the expanded text (e.g., the person calculating a similarity score based on a basic embeddings for the text and images); and
generating a summary document including the plurality of topics, the expanded text, and the selected image (e.g., by the person inserting the images within the summary).
This judicial exception is not integrated into a practical application because the additional elements of “at least one processor” (claim 13), “at least one memory” (claim 13), a “language generation model” (claims 1, 13 and 20) are all recited at a high-level of generality, and ¶[0084] of the Specification describes the use of a general-purpose processor, e.g., CPU. Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. In addition, the added limitation of using “a language generation model” (claims 1, 13 and 20) is not recited with sufficient specificity as to provide any details about how the language generation model operates or how the language generation model is anything than a tool for automating a processing which may be performed by the human mind. Thus, the claims as a whole are directed to an abstract idea (Step 2A, prong two).
Claims 1, 13 and 20 do not include any additional elements that are sufficient to amount to significantly more than the judicial exception because, as discussed above with respect to integration of the abstract idea into a practical application, the additional elements of “at least one processor” (claim 13), “at least one memory” (claim 13), a “language generation model” (claims 1, 13 and 20) amount to no more than mere instructions to apply the exception using generic computer components. See ¶[0084] which describes generic computer components and ¶¶[0001] and [00147] which describes use of conventional GPT models. Mere instructions to apply an exception using a generic computer component/model cannot provide an inventive concept (Step 2B).
With respect to dependent claims 2-4, 6, 8-12, 14, 15 and 19, these claims are directed to different elements, e.g., text, images, topics, which are extracted from the source document or summary . These limitations also relate to the abstract idea of “mental processes.” That is nothing in the claimed elements preclude the steps from practically being performed by a person using the recited criteria when grouping the document sentences. No additional elements are present.
With respect to dependent claims 5 and 7, these claims are directed generating text and image embedding for computing the similarity score and generating a synthesized image based on a topic. These limitations also relate to the abstract idea of “mental processes.” That is nothing in the claimed elements preclude the steps from practically being performed by a person generating a simple text embedding representing whether the text or image are related to each of the topics and/or the person drawing an image representative of a topic. No additional elements are present
With respect to dependent claims 16-18, these claims are directed also directed to the abstract idea, but recite the additional elements of “a user interface” (claim 16), “an image generation model” (claim 17), and “a transformer network” (claim 18). However, these additional elements do not (1) integrate the judicial exception into a practical application because they are all recited at a high-level of generality and are either known elements of a general purpose computer or commonly known learning models and (2) they are not sufficient to amount to significantly more than the judicial exception because they amount to no more than mere instructions to apply the judicial exception using generic computer components/models. Thus, these claims as a whole are directed to an abstract idea (Step 2A, prong two) and merely apply the exception using a generic/conventional computer/model which does not provide an inventive concept (Step 2B).
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 (i.e., changing from AIA to pre-AIA ) 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, 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 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 non-obviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 2, 4, 6, 8-9, 11-14, 16 and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Shahinian et al. (US 2024/0126981 A1; herein “Shahinian”), further in view of Wang et al. (US 2022/0269713 A; herein “Wang”) .
Regarding claims 1, 13, and 20, Shahinian teaches a method, an apparatus comprising at least one processor (¶[0009] teaches “The artificial intelligence assistant computing facility illustratively includes a processor…”) and at least one memory (¶[0009] teaches “…a memory in communication with the processor…”), and a non-transitory compute readable medium storing code, the code comprising instructions executable by at least one processor (¶[0009] teaches “…and some combination of logical modules for execution by the processor”) to perform the method comprising:
generating, using a language generation model, a summary of a source document (¶[0018] teaches “The illustrative method further includes generating, via the artificial intelligence assistant computing facility…a summary for each of the one or more presentations”; ¶[0028] teaches “The illustrative system is further configured to generate a summary for each of the one or more presentations”; ¶[0084] teaches “The artificial intelligence assistant computing facility 1000 also stores an AI-generated summary of the presentation”);
generating, using the language generation model, a plurality of topics based on the summary and a predetermined number of topics (¶[01000] teaches “The category classification module 1157 leverages a wide range of machine learning techniques to process content into relevant topics and subtopics…Relevant entities such as company, proper, and product names, as well as logos and other relevant content are classified for use in the auto-creation of slides and presentations” );
generating, using the language generation model, expanded text for each of the plurality of topics (¶[01000] teaches “…Relevant entities such as company, proper, and product names, as well as logos and other relevant content are classified for use in the auto-creation of slides and presentations” the auto-creation of slides and presentations is interpreted as the expansion of the topics and subtopics into text);
selecting an image from the source document for each of the plurality of topics by computing a similarity score between the image and the expanded text (¶[0107] teaches “The image suggestion module 1178 uses image recognition, neural nets, and machine learning to suggest images that are relevant to the text in a slide to use” relevant is interpreted as a similarity score); and
generating a summary document including the plurality of topics, the expanded text, and the selected image (¶[0098] teaches “The presentation auto-assembly module 1151 operates in the background to prepare the content for auto-assembly and includes a collection of document processors designed to work together to analyze, enrich, and tag content. The presentation auto-assembly module 1151 generates a new presentation based on topics and categories learned from existing presentations…and incorporates the best content associated with topics and categories requested by the user” and ¶[0094] teaches “The presentation editing module 1138 also handles and processes the user’s input to modify the text and images of content included in the generation of a new slide or presentation” Accordingly, the new presentation/slide include both text and images relevant to the user selected topic(s)).
Although Shahinian teaches generating new presentation slides based on a summary of source presentations, Shahinian fails to explicitly disclose that the new presentation is a summary of the source presentations.
Wang teaches a method and system for automatic generation of presentation slides from documents which includes an abstractive summary. More specifically, ¶{0007] teaches “The presentation slide generator is also configured to, when operating, generate an abstractive summary of the portions of the at least one source documents based on a long form questions answering machine learning process, and create a first presentation slide comprising the abstractive summary”. Thus, automatically generating slide presentations which include an abstractive summary of at least one source document is known in the art. Therefore, it would have been obvious to one having ordinary skill in the art, before the effective filing date of the invention, to have included generating presentation which include abstractive summaries of the source documents as taught by Wang in the system taught by Shahinian as it merely constitutes the combination of known processes to achieve the predictable results of providing a summary of the source documents that are most relevant to the user specified query and keywords (Wang, ¶[0024]).
Regarding claim 2, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above). In addition, Shahinian further teaches generating the summary comprises: extracting text content from the source document, wherein the summary is based on the text content (See Fig. 1, Slide content Extraction Module 1200 and Document extractor 1300; ¶[0007] teaches “The artificial intelligence assistant computing facility includes logic adapted to extract information from each slide” ).
Regarding claim 4, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above). In addition, Shahinian further teaches identifying a plurality of images from the source document and a pre-determined selection factor (¶[0110] teaches “The document extractor module 1300 identifies…and associated images from each slide. Furthermore, the document extractor module 1300 processes text and image from slides…” ); and
filtering the plurality of images based on the pre-determined selection factor to obtain a filtered set of images, wherein the filtered set of images includes the selected image (¶[0107] teaches “The image suggestion module 1178 uses image recognition, neural nets, and machine learning to suggest image that are relevant to the text in a slide to use. When the user selects image recommendations from a plurality of image suggestions, the image suggestion module incorporates the updates selected by the user to aid in tracking…and improving image recommendations” ).
Regarding claim 6, Shahinian teaches all of the elements of claim 1 (see detailed element mapping above). In addition, Shahinian further teaches extracting a plurality of images from the source document (¶[0110] teaches “The document extractor module 1300 identifies…and associated images from each slide. Furthermore, the document extractor module 1300 processes text and image from slides…”); and
filtering the plurality of images to obtain a filtered set of images, wherein the image is selected from the filtered set of images (¶[0107] teaches “The image suggestion module 1178 uses image recognition, neural nets, and machine learning to suggest image that are relevant to the text in a slide to use. When the user selects image recommendations from a plurality of image suggestions, the image suggestion module incorporates the updates selected by the user to aid in tracking…and improving image recommendations”).
Regarding claims 8 and 16, the combination of Shahinian and Wang teaches all of the elements of claims 1 and 13 (see detailed element mapping above). In addition, Shahinian further teaches wherein generating the plurality of topics comprises: generating a plurality of provisional topics (¶[0111] teaches “The document processor 1400 communicate with the AI engine 1600 to…classify topics for each document” );
receiving user input on the plurality of provisional topics (¶[0151] teaches “…the method includes enabling the user 1810 to like slides or add them to a list and suggesting slides for a new deck from liked slides or saved lists as well as all slides” ); and
updating the plurality of provisional topics based on the user input to obtain the plurality of topics (¶[0149] taches “For each slide in the new presentation, the system recommends text or images to modify the slide or auto-create a new slide to fit the new content and flow, as indicated in block 3025. The artificial intelligence assistant computing facility suggests text, images, or slides based on the skills learned by the AI Engine 1600 and from the input provided by the user 1810” Thus, suggest slides, which are generated based on the topics associated with the source presentations is updated based on the user interaction).
Regarding claim 9, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above). In addition, Shahinian further teaches wherein generating the plurality of topics comprises: generating a prompt for the language generation model that includes instructions to generate the plurality of topics to cover content of the summary (¶[0105] teaches “The AI summary module 1172 uses natural language generation to provide a summary of the content captured in the presentation deck. The AI summary module 1172 uses a text-to-text generation models to compile a concise summary of the presentation content” AI generation inherently includes providing a prompt to the AI engine/model. In addition, ¶[0111] teaches “The document processors 1400 communicate with the AI engine 1600 to … classify topics for each document”).
Regarding claims 11 and 19, the combination of Shahinian and Wang teaches all of the elements of claims 1 and 13 (see detailed element mapping above). In addition, Shahinian further teaches wherein generating the summary document comprises: obtaining a document template (¶[0133] teaches “The user 1810 can select a template 2130 to use for the new presentation or opt to accept the default template” and ¶[0078] teaches “During assembly, the presentation template determine either by a user or the system based on usage is leveraged in order to maintain the formatting and styling of every slide in a new auto-created presentation”); and
generating the summary document based on the document template (¶[0098] teaches “The presentation auto-assembly module 1151 incorporates slide templates and styling based on relevancy, recency, and popularity. The user can also select a specific template to be used for auto-generation.”).
Regarding claim 12, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above). In addition, Shahinian further teaches the summary document is a multi-modal handout including an image corresponding to each of the plurality of topics (¶[0013] teaches “The artificial intelligence assistant computing facility include logic to display text and image recommendations to tailor the slide based on an ensemble of machine learning models”).
Regarding claim 14, the combination of Shahinian and Wang teaches all of the elements of claim 13 (see detailed element mapping above). In addition, Shahinian further teaches the extraction component extracts text content and a plurality of images from the source document, wherein the summary is based on the text content (¶[0015] teaches “Illustratively, the artificial intelligence assistant computing facility includes a content auto-generation and organization module and an application user interface. This step is illustratively executed by a document extractor and a document processor of a slide content extraction module which, in turn, forms part of the content auto-generation and organization module”).
Claims 3 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Shahinian and Wang as applied to claims 2 and 14 above, and further in view of Chaturvedi et al. “Divide and Conquer: From Complexity to Simplicity for Lay Summarization” Proceeding of the First Workshop on Scholarly Document Processing; November 19, 2020, pages 344-355; herein “Chaturvedi.
Regarding claims 3 and 15, the combination of Shahinian and Wang teaches all of the elements of claims 2 and 14 (see detailed element mappings above). However, the combination of Shahinian and Wang fails to explicitly disclose generating the summary of the source document comprises: dividing the text content into a plurality of segments based on an input size of the language generation model; generating a plurality of segment summaries corresponding to the plurality of segments, respectively; and combining the plurality of segment summaries to obtain the summary of the source document.
Chaturvedi teaches a method generating lay summaries from complex scientific documents that includes, as shown in Fig. 1 on page 347, for all sections, with the exception of the abstract, of the source documents, first generating an extractive summarization of the section, then generating an abstractive summary from the extractive summarization. More specifically, Chaturvedi teaches generating the extractive summary of the source document comprises:
dividing the text content into a plurality of segments based on an input size of the language generation model (Page 345, section 1.3 teaches “We propose a two-step approach that divides the scientific scholarly text into segments…We discerningly combine state-of-the-art extractive…to first extract important sentences from the selected segments” and page 347, section 4.2 teaches “We over-determine important sentences from each of the remaining three sections using a common extractive summarization method…”);
generating a plurality of segment summaries corresponding to the plurality of segments, respectively ( Page 345, section 1.3 teaches “We propose a two-step approach that divides the scientific scholarly text into segments…We discerningly combine state-of-the-art extractive and abstractive summarization methods… and then compress and paraphrase these sentences via an abstractive summarizer” and page 347, section 4.2 teaches “…The four segments…are further condensed using an abstractive summarizer to obtain corresponding simplified text); and
combining the plurality of segment summaries to obtain the summary of the source document (Page 347, section 4.2 teaches “…The four segments…are further condensed using an abstractive summarizer to obtain corresponding simplified texts. Finally, the four abstractive summaries are concatenated one by one in the ACDI order until the desires Lay summary length is achieved.”).
The combination of Shahinian and Wang differs from the claimed invention, as defined by claims 3 and 15, in that the combination fails to explicitly disclose utilizing an extractive summarization on the source documents prior to generating the final multimodal summary. Utilizing an extractive summarization method prior to generating a final output summary is known in the art as evidenced by Chaturvedi. Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention, to have modified the summarization of the source presentations as taught by the combination of Shahinian and Wang to include an extractive summarization/abstractive summarization process as taught by Chaturvedi to (1) alleviate the shortcoming of transformer based models by reducing the input sequence and (2) improve the content selection in abstractive summaries (Chaturvedi, page 345, section 1.1).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over the combination of Shahinian and Wang as applied to claim 1 above, and further in view of Mukerjee et al. “Topic-aware Multimodal Summarization”, Findings of the Association for Computational Linguistics: AACL-IJCNLP 2022, pages 387-398; herein “Mukerjee”.
Regarding claim 5, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above).However, the combination of Shahinian and Wang fails to explicitly disclose generating a multi-modal text embedding based on the expanded text; generating a multi-modal image embedding based on the image; and computing the similarity score by comparing the multi-modal text embedding and the multi-modal image embedding.
Mukerjee teaches a topic-aware Multimodel Summarization method and system that includes, inter alia, generating a multi-modal text embedding based on the expanded text (Page 389 Fig. 2, reproduced below,Ctxt );
generating a multi-modal image embedding based on the image (Page 389 Fig. 2, reproduced below,Cimg); and
computing the similarity score by comparing the multi-modal text embedding and the multi-modal image embedding (Page 389 Fig. 2, reproduced below,Cmm Multimodal Attention Layer ).
PNG
media_image1.png
428
668
media_image1.png
Greyscale
Mukerjee Figure 2
The combination of Shahinian and Wang differs from the claimed invention, as defined by claim 5, in that the combination fails to explicitly is disclose utilizing using multimodal embeddings to select the relevant images. Utilizing multimodal embeddings to select topic relevant images is known in the art as evidenced by Mukherjee. Therefore, it would have been obvious to one having ordinary skill in the art, before the effective filing date of the invention, to utilized multimodal embeddings as taught by Mukherjee when generating the new presentations as taught by the combination of Shahinian and Wang as it merely constitutes the combination of known methods to achieve the predictable result of providing images that help users understand the text make the summary more attractive, contextualized, and complete (Mukherjee, page 387)
Claims 7, 10 and 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Shahinian and Wang as applied to claims 1 and 13 above, and further in view of Finegan et al. (US 11,797,780 B1; herein “Finegan”).
Regarding claims 7 and 17, the combination of Shahinian and Wang teaches all of the elements of claim 1 and 13 (see detailed element mapping above). However, the combination of Shahinian and Wang fails to explicitly disclose generating the summary document comprises: generating a synthesized image based on a topic of the plurality of topics, wherein the summary document includes the synthesized image.
Finegan teaches a method and system which utilizes a transformer architecture (200) to implement a text-to-image image model (118). More specifically, Finegan teaches generating a synthesized image based on a topic of the plurality of topics (Fig 3, steps 310-318 and Col 5, line 50 to col. 6, line 30 teaches “In block 310, as set of text documents is received…In block 312, a summary of the set of text documents is generated using a set of large language machine learning models…In block 314, a set of keywords is generated from the summary…reducing the summaries down to a number of keywords that grab the essence of the text documents and enabling the text-to-image model to generate images that are both relevant…”; col. 7, lines 4-9 teaches “In block 316, an image prompt is generate from the set of keywords…” and col. 7, lines 58-60 teaches “In block 318, a set of images is generated from the image prompt using a text-to-image machine learning model…”
The combination of Shahinian and Wang differs from the claimed invention as defined in claims 7 and 17, in that the combination fails to disclose that the images suggested for inclusion in the new presentation are synthesized images. Generating or synthesizing images relevant to a document summary is known in the art evidenced by Finegan. Therefore, it would have been obvious to one having ordinary skill in the art, before the effective filing date of the invention, to have modified the presentation generating system taught by Shahinian and Wang to include generating synthesized images as taught by Finegan as it merely discloses the combination of known processes to achieve the predicable result of providing suggested image which are generated by the image when the source document do not include sufficient or relevant images for the new presentation.
Regarding claim 10, the combination of Shahinian and Wang teaches all of the elements of claim 1 (see detailed element mapping above). However, the combination of Shahinian and Wang fails to explicitly teach generating the plurality of topics comprises: generating a prompt for the language generation model that includes instructions to generate the plurality of topics to be different from each other.
Finegan teaches a method and system which utilizes a transformer architecture (200) to implement a text-to-image image model (118) that includes generating a set of keywords from the summary and sentiment using a set of large language machine learning models. Specifically, col 6, lines 57-64 teaches “A clustering operation can then be performed to identify sentiments and/or topics, with each cluster of tokens representing a topic or sentiment…such as a short-text clustering algorithm that clusters the terms into a specified number, k, of clusters that form the sentiments and/or topics” the clusters are inherently different.
The combination of Shahinian and Wang differs from the claimed invention as defined in claim 10, in that the combination fails to disclose utilizing a large language model to generate a plurality of different topics. Generating a plurality of different topics utilizing a large language model is known in the art evidenced by Finegan. Therefore, it would have been obvious to one having ordinary skill in the art, before the effective filing date of the invention, to have modified the presentation generating system taught by Shahinian and Wang to utilize a large language model to generate a plurality of different clusters as taught by Finegan as it merely discloses the combination of known processes to achieve the predicable of preventing topic redundancy in the generated presentation.
Regarding claim 18, the combination of Shahinian and Wang teaches all of the elements of claim 13 (see detailed element mapping above). However, the combination of Shahinian and Wang fails to explicitly disclose the language generation model comprises a Transformer network.
Finegan teaches a method and system which utilizes a transformer architecture (200) to implement a text-to-image image model (118) that includes utilizing language generation model which is also a transformer. Specifically, col 3, lines 15-25 teaches that the language model maybe Open AI’s generative pretrained transformer 3 (GPT-3) model.
The combination of Shahinian and Wang differs from the claimed invention as defined in claim 18, in that the combination fails to disclose utilizing a transformer network. Generating or synthesizing text summaries and images using a transformer network is known in the art evidenced by Finegan. Therefore, it would have been obvious to one having ordinary skill in the art, before the effective filing date of the invention, to have modified the presentation generating system taught by Shahinian and Wang to utilize a transformer based language model as taught by Finegan as it merely discloses the combination of known processes to achieve the predicable result of generating summaries and corresponding topics included therein.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PENNY L CAUDLE whose telephone number is (703)756-1432. The examiner can normally be reached M-Th 8:00 am to 5:00 pm eastern.
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, Daniel Washburn can be reached at 571-272-5551. 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.
/PENNY L CAUDLE/Examiner, Art Unit 2657