Patent Application 18356116 - DOMAIN SPECIALTY INSTRUCTION GENERATION FOR TEXT

Title: DOMAIN SPECIALTY INSTRUCTION GENERATION FOR TEXT ANALYSIS TASKS
Application Information

Invention Title: DOMAIN SPECIALTY INSTRUCTION GENERATION FOR TEXT ANALYSIS TASKS
Application Number: 18356116
Submission Date: 2025-05-13T00:00:00.000Z
Effective Filing Date: 2023-07-20T00:00:00.000Z
Filing Date: 2023-07-20T00:00:00.000Z
National Class: 704
National Sub-Class: 235000
Examiner Employee Number: 84589
Art Unit: 2692
Tech Center: 2600
Rejection Summary

102 Rejections: 0
103 Rejections: 1
Cited Patents

No patents were cited in this rejection.
Office Action Text



    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 .

DETAILED ACTION

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 Krishna: 20190325066 hereinafter Kri further in view of Smith: 20240273291.

Regarding claim 1
Kri teaches:
A system, method, etc. operative upon one or more computing devices, respectively comprising at least one processor and a memory; wherein the one or more computing devices store program instructions (Kri: Abstract; Fig 1, 9: that when executed by the one or more computing devices: 
receive a request to perform a summarization task on a natural language text, wherein the request specifies a domain specialty (Kri: ¶ 70-74: system receives request to generate topic based summary of text with respect to a topic of interest the request including textual content and topic of interest); 
insert one or more domain specialty identifiers as part of generating instructions to perform the summarization task using a pre-trained language model fine-tuned to a domain comprising a plurality of domain specialties including the domain specialty (Kri: ¶ 65-68, 70-74; Fig 7, 8: textual content encoded by selection of a topic vector and encoding the textual content by mapping words to a topic vector which encodes a topic with each determined word; said encoding, mapping, etc. based on previously trained topic aware encoding models operable to generate topic based summaries of an input; that is, the system generates instructions for selecting each/any next word of a summary in concert with an attention distribution); 
cause the pre-trained language model fine-tuned to the domain to perform the summarization task on the natural language text using the generated instructions (Kri: ¶ 65-68, 70-74; Fig 7, 8: system performs summarization task based on instruction relevant to the selection of each/any next word of the summary); and 
provide a result of the summarization task performed on the natural language text (Kri: ¶ 27, 65-68, 70-74; Fig 1, 7, 8: system provides a topic based summary of the input text tuned to the topic of interest).

As such Kri at least strongly suggests generating instructions for the performance of summarization but does not explicitly teach a system operative to perform the summarization task using a pre-trained large language model.

In a related field of endeavor Smith teaches a system and method operable for outputting, publishing, etc. summaries of articles, documents, texts, etc. (Smith: Abstract: ¶ 35, 37, 128, etc.: system generates writing based on an article, in response to a prompt) said summaries generated by a pre trained large language model (LLM—Smith:¶ 54, 70: output generated using predictive models such as GPT, BERT, etc.); the system additionally operative to generate the output by inserting one or more domain specialty identifiers as part of generating instructions to perform the summarization (Smith: ¶ 182-185: a prompt generator comprises a template for inputting instructions relevant to a determined domain). It would have been obvious to one of ordinary skill in the art before the effective filing date of the instant application to adapt the Kri domain based summarization system to utilize a pretrained large language model and to include the additional generating instructions taught or suggested by Smith for at least the purpose of generating summaries of large corpora of texts; one of ordinary skill in the art would have expected only predictable results from such an adaptation.

Regarding claim 2
Kri in view of Smith teaches or suggests:
The system of claim 1, wherein the one or more computing devices store further program instructions that when executed by the one or more computing devices insert a definition of the domain specialty as part of generating the instructions to perform the summarization task using the pre-trained large language model fine-tuned to the domain (Kri: ¶ 26-28, 35-46; claim 1, 9, 10: system generates a summary based on a request defining the topic of interest which resolves topic definitions across a corpora; based on the specification as filed a definition is given its plain meaning; that of a statement of bounds or limits and as such the broadest reasonable interpretation is considered as substantially similar to specifying a domain, domain specialty identifier, etc. as stated in claim 1 and as represented in Kri as that model and dataset being tuned to tagging or labelling a document and summary with a topic or topics generative of topic based groups); (Smith: 86, 1146, 82-185: a defined domain adapted to a prompt template; the definition is considered additional applicable to a topic and subtopics such as represented by concepts I of attendant nodes in a knowledge graph and resolvable by a topic descriptor, keyword, sets thereof as represented by a user prompt). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 3
Kri in view of Smith teaches or suggests:
The system of claim 1, wherein the one or more computing devices store further program instructions that when executed by the one or more computing devices generate the natural language text as a transcript from obtained audio data using an automatic speech recognition system (Smith: ¶ 139: a speech recognition system applied to an audio input to generate a text transcription for generating a subsequent summary, textual output, etc.). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 4
Kri in view of Smith teaches or suggests:
The system of claim 1, wherein the one or more computing devices are implemented as part of a medical audio summarization service offered as part of a provider network and wherein the request is received via an interface of the medical audio summarization service. Examiner takes official notice that the utility of summarization such as on patient, medical etc. data; said data comprising input audio, transcripts thereof, textual matter related thereto was an extremely well known utility in the art before the effective filing date of the instant invention and would have comprised an obvious utility of the Kri in view of Smith system and method for at least the purpose of automating procedures related to patient care and medial business management; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 5, 14—the claims recite substantially similar subject matter to that of claim 1 and are similarly rejected.

Regarding claim 6, 15—the claims recite substantially similar subject matter to that of claim  and are similarly rejected.

Regarding claim 7
Kri in view of Smith teaches or suggests:
The method of claim 5, wherein the text analysis system supports a plurality of different domains, including the domain, and wherein the domain is specified as part of a request to perform the text analysis task (Kri: Abstract; ¶ 18, 22, 40-49; Fig 4, 5: system operative to generate plural summaries with respect to plural domains, topics, etc.); (Smith: system utilizes a model operative to address multiple domains, topics, etc.). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 8
Kri in view of Smith teaches or suggests:
The method of claim 5, wherein the text analysis task is a summarization task (Kri: ¶ 27, 65-68, 70-74; Fig 1, 7, 8: system provides a topic based summary of the input text tuned to the topic of interest); (Smith: Abstract: ¶ 35, 37, 128, etc.: system generates writing based on an article, in response to a prompt). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 9, 16—the claims recite substantially similar subject matter to that of claim 3 and are similarly rejected.

Regarding claim 10
Kri in view of Smith teaches or suggests:
The method of claim 5, further comprising: receiving a request to add a new specialty to the domain (Kri: ¶ 40-43; Fig 4, 5; claim 1, 9, 10: such as by addition of new, second, subsequent topics to the training data in concert with a domain definition; such as by iteration of the figure 4 method), wherein the request includes a definition for the new specialty (Kri: ¶ 40-43; Fig 4, 5; claim 1, 9, 10: system generates a summary based on a request defining the topic of interest which resolves topic definitions across a corpora); and performing further fine-tuning on the pre-trained large language model for the domain using additional training data annotated with specialty identifiers for the new specialty (Kri: ¶ 40-43; Fig 4, 5; claim 1, 9, 10: such as by further training with adapted training data resultant from iterating the figure 4 method); (Smith: ¶ 170, 332: such as by identifying and incorporating new applications or domains in the model). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 11
Kri in view of Smith teaches or suggests:
The method of claim 5, wherein the domain specialty is identified for a plurality of different texts, including the input text as part of performing batch processing of the text analysis task on the plurality of different texts (Kri: ¶ 34-36, etc.; Fig 2: plural topics labeled within each/any of a corpus of documents including the input text; this comprises batch processing in as much as multiple text files are processed similarly and at a substantially similarly time and used for performing the same text analysis task); (Smith: ¶ 170, 332 such as by identifying and incorporating new applications or domains in the model; such as by batch processing of the data store in an offline manner). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.

Regarding claim 12
Kri in view of Smith teaches or suggests:
The method of claim 5, wherein identifying the domain specialty for the input text comprises performing an entity recognition technique on the input text (Smith: ¶ 84-87; Fig 2, etc.: system operates to determine, describe, etc. entities in a system space such as for an input text, corpus thereof, etc.).

Regarding claim 13
Kri in view of Smith teaches or suggests:
The method of claim 5, wherein identifying the domain specialty for the input text comprises identifying the domain specialty as specified in a request to perform the text analysis task (Kri: ¶ 70-74: system receives request to generate topic based summary of text with respect to a topic of interest the request including textual content and topic of interest). The claim is considered obvious over Kri as modified by Smith as addressed in the base claim as it would have been obvious to apply the further teaching of Kri and/or Smith to the modified device, method, etc. of Kri and Smith; one of ordinary skill in the art would have expected only predictable results therefrom.
   
Regarding claim 17—the claim recites substantially similar subject matter to that of claim 11 and is similarly rejected.
   
Regarding claim 18—the claim recites substantially similar subject matter to that of claim 12 and is similarly rejected.
   
Regarding claim 19—the claim recites substantially similar subject matter to that of claim 13 and is similarly rejected.
   
Regarding claim 20—the claim recites substantially similar subject matter to that of claim 4 and is similarly rejected.

Conclusion
 
The  
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