DETAILED ACTION
Claim Status
This is first office action on the merits in response to the application filed on 5/21/2024.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-17 are currently pending and have been examined.
Claim Objections
Claims 2, 8, 10, and 16 are objected to because of the following informalities:
In claim 2, line 2; and claim 10, line 2, “an output” should read --the output--.
In claim 8, line 2; and claim 16, line 3, “the body” should read --a body--.
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-17 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Under the Step 1 of the Section 101 analysis, Claims 1-8 are drawn to a method which is within the four statutory categories (i.e., a process), Claims 9-16 are drawn to a system which is within the four statutory categories (i.e. a machine), and Claim 17 is drawn to a non-transitory computer-readable medium which is within the four statutory categories (i.e., a manufacture).
Since the claims are directed toward statutory categories, it must be determined if the claims are directed towards a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea). Based on consideration of all of the relevant factors with respect to the claim as a whole, claims 1-17 are determined to be directed to an abstract idea. The rationale for this determination is explained below:
Regarding Claims 1, 9, and 17:
Claims 1, 9, and 17 are drawn to an abstract idea without significantly more. The claims recite “receiving a message; generating a prompt for an LLM including the message and an instruction to generate an output identifying predetermined content in the message; passing the prompt through an LLM to generate the output; processing the output in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated.”
Under the Step 2A Prong One, the limitations, as underlined above, are processes that, under its broadest reasonable interpretation, cover Mental Processes such as concepts performed in the human mind (including an observation, evaluation, judgment, opinion).
For example, but for the “LLM”, “hallucination detection process”, and “hallucinated” language, the underlined limitations in the context of this claim encompass the human activity or mental processes. A person could receive a message. The person could then generate and pass a prompt. Finally, the person could process output to identify if a predetermined content is hallucinated, as in a typical human mind.
Under the Step 2A Prong Two, this judicial exception is not integrated into a practical application. In particular, the claim only recites additional elements – “A computer implemented method of detecting hallucination in a large language model (LLM) output, said method comprising the steps of:”, “A computer system for detecting hallucination in a large language model (LLM) output, said system comprising a message receiving module communicatively connected to a prompt generation module, … a hallucination detection module”, “A computer program which when executed on a computing device controls the computing device to implement a method”, “LLM”, “hallucination detection process”, and “hallucinated”. The additional elements are recited at a high-level of generality (i.e., performing generic functions of an interaction) such that it amounts no more than mere instructions to apply the exception using a generic computer component, merely implementing an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea. Additionally, regarding the specification and claims, there is no improvement in the functioning of a computer or an improvement to other technology or technical field present, there is no applying or using the judicial exception to effect a particular treatment or prophylaxis for a disease or medical condition present, there is no implementing the judicial exception with or using the judicial exception in conjunction with a particular machine or manufacture that is integral to the claim present, there is no effecting a transformation or reduction of a particular article to a different state or thing present, and there is no applying or using the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment present such that the claim as a whole is more than a drafting effort designed to monopolize the exception. Accordingly, these additional elements, individually or in combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Under the Step 2B, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements in the process amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claims are not patent eligible.
Regarding Claims 2-8 and 10-16:
Dependent claims 2, 5-6, 10, and 13-14 only further elaborate the abstract idea and do not recite additional elements.
Dependent claims 3-4, 7-8, 11-12, and 15-16 include additional limitations, for example, “LLM” (Claims 3 and 11); “LLM” and “LLM API” (Claims 4 and 12); “email” (Claims 7 and 15); and “email” (Claims 8 and 16), but none of these limitations are deemed significantly more than the abstract idea because, as stated above, they require no more than generic computer structures or signals to be executed, and do not recite any Improvements to the functioning of a computer, or Improvements to any other technology or technical field.
Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Furthermore, looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology, and their collective functions merely provide conventional computer implementation or implementing the judicial exception on a generic computer.
Therefore, whether taken individually or as an ordered combination, claims 1-17 are nonetheless rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
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.
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.
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 nonobviousness.
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) 1-7, 9-15, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sheth (US 20240370769 A1) in view of Kumar (US 20240330755 A1).
Regarding Claims 1, 9, and 17, Sheth teaches A computer implemented method of detecting hallucination in a large language model (LLM) output, said method comprising the steps of (Sheth: Abstract; Paragraph(s) 0019-0025): A computer system for detecting hallucination in a large language model (LLM) output, said system comprising a message receiving module communicatively connected to a prompt generation module, … a hallucination detection module… (Sheth: Abstract; Paragraph(s) 0019-0025): A computer program which when executed on a computing device controls the computing device to implement a method (Sheth: Abstract; Paragraph(s) 0019-0025),
receiving a message (Sheth: Paragraph(s) 0006-0007 teach(es) the hallucination prevention system may receive from the specialized language model, a plurality of responses to the plurality of sub-queries);
generating a prompt for an LLM including the message and an instruction to generate an output identifying predetermined content in the message (Sheth: Paragraph(s) 0007, 0027, 0031, 0041-0044, 0047-0048, 0050-0053 teach(es) The hallucination prevention system may then generate an engineered prompt for a machine learning model (e.g., a large language model));
passing the prompt through an LLM to generate the output (Sheth: Paragraph(s) 0007 teach(es) The hallucination prevention system may then input the engineered prompt into the large language model to generate the response based on the “facts” it is supplied).
However, Sheth does not explicitly teach processing the output in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated.
Kumar from same or similar field of endeavor teaches processing the output in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated (Kumar: Abstract; Paragraph(s) 0065, 0078-0080 teach(es) To determine whether a resolution statement being analyzed includes hallucinated content or not, the hallucination detector may be configured to generate a hallucination score for the statement, where the score may be generated as a weighted sum of hallucination factors, also referred to as factors).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Sheth to incorporate the teachings of Kumar for processing the output in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated.
There is motivation to combine Kumar into Sheth because Kumar’s teachings of hallucination detector would facilitate to remove the hallucinated content (Kumar: Abstract; Paragraph(s) 0065).
Regarding Claims 2 and 10, the combination of Sheth and Kumar teaches all the limitations of claims 1 and 9 above; however the combination does not explicitly teach further comprising: generating an output indicative of whether the hallucination detection process has identified that the predetermined content identified in the output is potentially hallucinated.
Kumar further teaches further comprising: generating an output indicative of whether the hallucination detection process has identified that the predetermined content identified in the output is potentially hallucinated (Kumar: Abstract; Paragraph(s) 0065, 0078-0080, as stated above with respect to claims 1 and 9).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Kumar for processing the output in accordance with a hallucination detection process to identify if any predetermined content identified by the LLM in the output is potentially hallucinated.
There is motivation to combine Kumar into the combination of Sheth and Kumar because Kumar’s teachings of hallucination detector would facilitate to remove the hallucinated content (Kumar: Abstract; Paragraph(s) 0065).
Regarding Claims 3 and 11, the combination of Sheth and Kumar teaches all the limitations of claims 1 and 9 above; however the combination does not explicitly teach processing the output in accordance with the hallucination detection process comprises: performing a content search operation of the received message to identify in the received message the predetermined content identified in the output of the LLM, and, if the predetermined content identified in the output of the LLM is not identified in the received message, identifying the output of the LLM as potentially hallucinated.
Kumar further teaches processing the output in accordance with the hallucination detection process comprises: performing a content search operation of the received message to identify in the received message the predetermined content identified in the output of the LLM, and, if the predetermined content identified in the output of the LLM is not identified in the received message, identifying the output of the LLM as potentially hallucinated (Kumar: Paragraph(s) 0108-0109, 0113, 0115-0116 teach(es) the dynamic source context may be determined using other techniques, such as search techniques or resolution insights; Within the dynamic source context, multiple incident tickets may be included, and one or more sentences from within one or more incident ticket(s) that is or are maximally similar to the input text may be used to compute a hallucination score).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Kumar for processing the output in accordance with the hallucination detection process comprises: performing a content search operation of the received message to identify in the received message the predetermined content identified in the output of the LLM, and, if the predetermined content identified in the output of the LLM is not identified in the received message, identifying the output of the LLM as potentially hallucinated.
There is motivation to combine Kumar into the combination of Sheth and Kumar because Kumar’s teachings of search techniques would facilitate to find out the hallucinated content (Kumar: Abstract; Paragraph(s) 0108-0109, 0113, 0115-0116).
Regarding Claims 4 and 12, the combination of Sheth and Kumar teaches all the limitations of claims 1 and 9 above; however the combination does not explicitly teach processing the output in accordance with the hallucination detection process comprises: receiving from an LLM API confidence score data associated with the output generated by the LLM; determining if a confidence score associated with the confidence score data exceeds a predetermined confidence threshold, and, if the confidence score does not exceed the predetermined threshold, identifying the output of the LLM as potentially hallucinated.
Kumar further teaches processing the output in accordance with the hallucination detection process comprises: receiving from an LLM API confidence score data associated with the output generated by the LLM; determining if a confidence score associated with the confidence score data exceeds a predetermined confidence threshold, and, if the confidence score does not exceed the predetermined threshold, identifying the output of the LLM as potentially hallucinated (Kumar: Abstract; Paragraph(s) 0078-0079 teach(es) The score generator may use the determined weights and factors of the hallucination equation to generate a composite hallucination score for a resolution statement. In some implementations, the hallucination score may be compared to a threshold value).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Kumar for processing the output in accordance with the hallucination detection process comprises: receiving from an LLM API confidence score data associated with the output generated by the LLM; determining if a confidence score associated with the confidence score data exceeds a predetermined confidence threshold, and, if the confidence score does not exceed the predetermined threshold, identifying the output of the LLM as potentially hallucinated.
There is motivation to combine Kumar into the combination of Sheth and Kumar because Kumar’s teachings of composite hallucination score and threshold would facilitate to find out the hallucinated content (Kumar: Abstract; Paragraph(s) 0078-0079).
Regarding Claims 5 and 13, the combination of Sheth and Kumar teaches all the limitations of claims 1 and 9 above; however the combination does not explicitly teach wherein the predetermined content comprises predetermined business process metadata.
Kumar further teaches wherein the predetermined content comprises predetermined business process metadata (Kumar: Paragraph(s) 0049-0050, 0102, 0104, 0056-0058 teach(es) A domain metadata generator may be configured to identify metadata characterizing incident tickets in the ticket data repository, which may then be included in the training analysis results. Domain metadata may thus include any field or other characteristic of an incident ticket(s) a, including, e.g., related services or products; FIG. 5 illustrates an example implementation of the hallucination detection and correction system for domain-specific machine learning models of FIG. 1 when a resolution is generated from within an input source. In FIG. 5, a table 502 represents a plurality of incident tickets a of FIG. 1 and associated data and metadata; By populating the training analysis results with domain vocabulary, domain metadata, generated clusters, and relevant tuning parameters, the training analyzer enables the hallucination detector to detect hallucinated outputs provided by the LLM).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Kumar for wherein the predetermined content comprises predetermined business process metadata.
There is motivation to combine Kumar into the combination of Sheth and Kumar because Kumar’s teachings of domain metadata would facilitate detect hallucinated outputs provided by the LLM (Kumar: Abstract; Paragraph(s) 0049-0050, 0102, 0056).
Regarding Claims 6 and 14, the combination of Sheth and Kumar teaches all the limitations of claims 5 and 13 above; and Sheth further teaches wherein the predetermined business process metadata comprises predetermined financial transaction metadata (Sheth: Paragraph(s) 0039-0040 teach(es) Some other examples for a corpus of information may include, financial statements, medical journals, etc.).
Regarding Claims 7 and 15, the combination of Sheth and Kumar teaches all the limitations of claims 1 and 9 above; however the combination does not explicitly teach wherein the message is an email.
Kumar further teaches wherein the message is an email (Kumar: Paragraph(s) 0033, 0045-0046 teach(es) The worklog may include attempted resolutions performed by the incident agent, messages (e.g., emails or chat messages) between the user and the incident agent, or written, recorded-audio, or auto-transcribed text of audio communications between the user and the incident agent).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Kumar for wherein the message is an email.
There is motivation to combine Kumar into the combination of Sheth and Kumar because Kumar’s teachings of email messages would facilitate hallucination detection (Kumar: Paragraph(s) 0033, 0045-0046).
Claim(s) 8 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sheth (US 20240370769 A1) in view of Kumar (US 20240330755 A1), as applied to claims 7 and 15 above, and in further view of Lutz (US 20240311348 A1).
Regarding Claims 8 and 16, the combination of Sheth and Kumar teaches all the limitations of claims 7 and 15 above; however the combination does not explicitly teach wherein generating the prompt for the LLM comprises generating a prompt including unstructured text data from the body and/or header of the email and the instruction to generate the output identifying predetermined content in the text data of the message.
Lutz from same or similar field of endeavor teaches wherein generating the prompt for the LLM comprises generating a prompt including unstructured text data from the body and/or header of the email and the instruction to generate the output identifying predetermined content in the text data of the message (Lutz: Paragraph(s) 0096-0098, 0093 teach(es) FIG. 8 shows a case in which the context-bearing portion is clearly demarcated by the user from the query portion; the application management component instructs the creation system to create the structured database based on the unstructured text of the context-bearing portion; the creation system advances through the Email messages sentence-by-sentence).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of the combination of Sheth and Kumar to incorporate the teachings of Lutz for wherein generating the prompt for the LLM comprises generating a prompt including unstructured text data from the body and/or header of the email and the instruction to generate the output identifying predetermined content in the text data of the message.
There is motivation to combine Lutz into the combination of Sheth and Kumar because Lutz’s teachings of processing of unstructured text including a body of email would facilitate to reduce the risk of artificial hallucination (Lutz: Paragraph(s) 0096-0098, 0093).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Fabian (US 20240303440 A1) teaches Prompt Configuration For LLM Integrations In Spreadsheet Environments, including detect hallucinated references and confidence.
Imani (US 20240362417 A1) teaches Readability Based Confidence Score For Large Language Models, including confidence score and hallucinating.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CLAY LEE whose telephone number is (571)272-3309. The examiner can normally be reached Monday-Friday 8-5pm EST.
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, Neha Patel can be reached at (571)270-1492. 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.
/CLAY C LEE/Primary Examiner, Art Unit 3699
Prosecution Insights