DETAILED ACTION
This Non-Final communication is in response to Application No. 18/621,796 filed 3/29/2024. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Claims 1-20 have been examined.
Claim Objections
Claim 18 is objected to because of the following informalities: it appears “…the second generative AI model…” in line 20 was intended. Appropriate correction is required.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1, 2, and 8 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Hernandez et al. (US 2025/0265413 A1, filed 2/16/2024).
Regarding claim 1, Hernandez teaches a computer-implemented method for capturing a dynamic state of a thread, the method comprising:
receiving, at an application, a first query for a thread; receiving, at the application, a first response to the first query. More specifically, customers initiate chat sessions with agents; the customers can submit queries, and the agents return responses or vice versa (Hernandez, [0039]-[0042]).
generating a first prompt wherein the first prompt includes the first query and the first response for the first query; providing the first prompt to a generative artificial intelligence (AI) model. More specifically, customer input data can be one message and agent input data can be one message (query-response pair) and is associated with a first state. Data (prompt) is generated with the customer input data and agent input data (query-response) and sent to a large language model for generating a summary (descriptor) (Hernandez, [0046]).
receiving, from the generative AI model in response to the first prompt, a first thread descriptor representing a first state of the thread; surfacing the first thread descriptor. More specifically, a first summary (thread descriptor) is generated with the large language model and the agents chat interface is updated with the first summary (Hernandez, abstract, [0046]).
receiving, at the application, a second query for the thread and a second response to the second query. More specifically, at a second state (a second query and second response can be used) (Hernandez, abstract, [0065]).
generating a second prompt wherein the second prompt includes the second query and the second response for the second query; providing the second prompt to the generative AI model. More specifically, the large language model is sent a prompt with the second customer message query and second agent message (second query-response) (Hernandez, abstract, [0065]).
receiving, from the generative AI model in response to the second prompt, a second thread descriptor, representing a second state of the thread; and surfacing the second thread descriptor, wherein the second thread descriptor replaces the first thread descriptor. More specifically, the large language model outputs a second summary (second thread descriptor) and updates the agents chat interface with the second summary (Hernandez, abstract, [0065]).
Regarding claim 2, Hernandez teaches the computer-implemented method of claim 1, wherein the application is a web browser having a chat interface. More specifically, the chat session is executed in a web browser (Hernandez, [0035]-[0038])
Regarding claim 8, Hernandez teaches the computer-implemented method of claim 1, wherein the thread is a first thread and the method further comprises: storing the second thread descriptor with the first thread; receiving, at the application, a first query for a second thread; generating a third thread descriptor, representing a state of the second thread, based on at least one of the first query of the second thread or a first response to the first query for the second thread; and storing the third thread descriptor with the second thread. More specifically, Figure 5B depicts a display depicting two conversations and their summaries generated by the process cited above (Hernandez, Figure 5B, [0109]).
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) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez, and further in view of Sotiriou et al. (US 2025/0069086 A1, filed 2/14/2024, hereinafter “Sotiriou”).
Regarding claim 3, Hernandez teaches the computer-implemented method of claim 1, however, may not explicitly teach every aspect of wherein the first prompt further includes grounding data used in generating the first response.
Sotiriou discloses a method for enhancing agent-customer interactions in a digital engagement service. The method includes receiving a customer's communication request (query) and presenting the request to an available agent through a user interface for generating a response (Sotiriou, abstract). The process of getting a response from an agent includes generating a prompt to an LLM which includes at least previous chatbot interactions (query context) and customer profile data (grounding data) and presenting the result to the agent (Sotiriou, [0065], [0081]). After a response/discussion is provided by the agent in step 408, another prompt is generated to the LLM at 414 for generating a summary of the conversation between the customer and the agent (Sotiriou, [0069]-[0070]). The prompt for the summarization of the conversation uses both the customer interactions as well as context/profile data (grounding data) (Sotiriou, [0078], [0087], [0089])
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Sotiriou that a method for generating a descriptor of a thread of query-response pairs using generated prompts would include using the grounding data used in generating the responses for the prompt for generating the descriptor. With Hernandez and Sotiriou disclosing interaction between a client and agent involving queries and responses, and prompting a generative AI model for generating summaries of the interaction, and with Sotiriou additionally disclosing using the same contextual (grounding) data when generating responses as well as the interaction summaries, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of query-response pairs using generated prompts would include using the grounding data used in generating the responses for the prompt for generating the descriptor in order to ensure the responses and summaries are the most relevant to the queries. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of query-response pairs using generated prompts.
Claim(s) 4 and 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez, and further in view of Tanikella (US 2023/0376515 A1, filed 5/18/2022).
Regarding claim 4, Hernandez teaches the computer-implemented method of claim 1, however, may not explicitly teach every aspect of wherein the second thread descriptor includes at least one of a title for the thread, a synopsis for the thread, or an image for the thread.
Tanikella discloses techniques for generating summary documents from conversations within a communication platform. A machine-learned summarization model can receive some or all of the contents of a virtual space and can generate a summarization document reflecting the contents of various communications (Tanikella, abstract). The summary document may include various visualizations (images), such as but not limited to, a cluster of profile photo of icons representing engagements of users in the conversation, a cluster of emojis used in the conversations representing the sentiments of the conversation, a compilation of visual elements in a tree structure representing the intensity and evolution of user reactions and engagements in the conversation, and/or the like (Tanikella, [0022]). The communication platform can treat the summary document as a synopsis or outline of the virtual space and the posts, events, etc. that took place within the virtual space (Tanikella, [0023]). The summary can consist of a title (Tanikella, [0129]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Tanikella that a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is one of a title, synopsis, or an image. With Hernandez and Tanikella disclosing generating summaries of communications, and with Tanikella additionally disclosing the summaries include one of a title, synopsis, or an visual/image, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is one of a title, synopsis, or an image in order to provide a more user friendly way of distinguishing the summaries from each other. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
Regarding claim 5, Hernandez and Tanikella teach the computer-implemented method of claim 4, wherein the second thread descriptor is the synopsis for the thread, and the second prompt includes static instructions for generating the synopsis. More specifically, Tanikella suggests the descriptor is one of a title, synopsis, or an image (Tanikella, [0022], [0023], [0129]). Hernandez suggests static instructions for the summary prompt can include a randomness parameter, minimum/maximum length, etc. (Hernandez, [0103]).
Claim(s) 6 and 7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez and Tanikella, and further in view of Wilson et al. (US 2022/0068296 A1, hereinafter “Wilson”).
Regarding claim 6, Hernandez and Tanikella teach the computer-implemented method of claim 4, wherein the second thread descriptor is the image for the thread. More specifically, Tanikella suggests the descriptor is one of a title, synopsis, or an image (Tanikella, [0022], [0023], [0129]). Hernandez additionally suggests static instructions for the summary prompt can include a randomness/creativeness parameter, minimum/maximum length, etc. (Hernandez, [0103]).
However, Hernandez and Tanikella may not explicitly teach every aspect of
the second prompt includes static instructions for generating the image.
Wilson discloses an approach to generating image representations of a conversation between a plurality of users. Generating one or more image representations of the one or more detected utterances utilizes a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. Embodiments of the present invention summarize conversations and display textual topics that are difficult to convey without visual cues. The program 150 utilizes the identified information as inputs to a trained GAN (i.e., image generator model 156) creating image representations of conversations (Wilson, abstract, [0010], [0023], [0037], at least user preferences or privacy parameters can be construed as static instructions for generating the image). The conversations are any of email, instant message, direct message, text message, social media post, spoken message, etc. (Wilson, [0026]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Tanikella with Wilson that a method for generating a descriptor of a thread of communications using generated prompts, the descriptor being one of a title, synopsis, or an image would include that generating an image representation of the descriptor uses a prompt with static instructions for generating the image. With Hernandez, Tanikella, and Wilson disclosing generating summaries of conversations, with Tanikella and Wilson disclosing the summaries including an visual/image, and with Wilson additionally disclosing using static instructions for generating the image, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts, the descriptor being one of a title, synopsis, or an image would include that generating an image representation of the descriptor uses a prompt with static instructions for generating the image in order to allow a user to set and maintain their privacy in image generation technologies. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
Regarding claim 7, Hernandez teaches the computer-implemented method of claim 1, however, may not explicitly teach every aspect of wherein the second thread descriptor includes at least one of a synopsis or title for the thread.
Tanikella discloses techniques for generating summary documents from conversations within a communication platform. A machine-learned summarization model can receive some or all of the contents of a virtual space and can generate a summarization document reflecting the contents of various communications (Tanikella, abstract). The summary document may include various visualizations, such as but not limited to, a cluster of profile photo of icons representing engagements of users in the conversation, a cluster of emojis used in the conversations representing the sentiments of the conversation, a compilation of visual elements in a tree structure representing the intensity and evolution of user reactions and engagements in the conversation, and/or the like (Tanikella, [0022]). The communication platform can treat the summary document as a synopsis or outline of the virtual space and the posts, events, etc. that took place within the virtual space (Tanikella, [0023]). The summary can consist of a title (Tanikella, [0129]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Tanikella that a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is one of a title or synopsis. With Hernandez and Tanikella disclosing generating summaries of communications, and with Tanikella additionally disclosing the summaries include one of a title, synopsis, or an visual/image, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is one of a title or synopsis in order to provide a more user friendly way of distinguishing the summaries from each other. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
While Tanikella additionally suggests the summary includes generated images, Hernandez and Tanikella may not explicitly teach every aspect of
and the method further comprises: generating a third prompt including at least the synopsis or title and static instructions for generating an image; providing the third prompt to the generative AI model; and receiving, from the generative AI model, the image as another thread descriptor.
Wilson discloses an approach to generating image representations of a conversation between a plurality of users. Generating one or more image representations of the one or more detected utterances utilizes a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. Embodiments of the present invention summarize conversations and display textual topics that are difficult to convey without visual cues. The program 150 utilizes the identified information as inputs to a trained GAN (i.e., image generator model 156) creating image representations of conversations (Wilson, abstract, [0010], [0023], [0037], at least user preferences or privacy parameters can be construed as static instructions for generating the image). The conversations are any of email, instant message, direct message, text message, social media post, spoken message, etc. (Wilson, [0026]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Tanikella with Wilson that a method for generating a descriptor of a thread of communications using generated prompts, the descriptor being one of a title, synopsis, or an image would include that generating an image representation of the descriptor uses a prompt with static instructions for generating the image. With Hernandez, Tanikella, and Wilson disclosing generating summaries of conversations, with Tanikella and Wilson disclosing the summaries including an visual/image, and with Wilson additionally disclosing using static instructions for generating the image, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts, the descriptor being one of a title, synopsis, or an image would include that generating an image representation of the descriptor uses a prompt with static instructions for generating the image in order to allow a user to set and maintain their privacy in image generation technologies. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
Claim(s) 9-12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez, and further in view of Han et al. (US 2020/0059548 A1, hereinafter “Han”).
Regarding claim 9, Hernandez teaches the computer-implemented method of claim 8, however, may not explicitly teach every aspect of further comprising: generating a user interface (UI) comprising: a first selectable UI element representing the first thread, the first selectable UI element including the second thread descriptor; and a second selectable UI element representing the second thread, the second selectable UI element including the third thread descriptor.
Han discloses an artificial intelligence unit for acquiring at least one conversation, acquiring at least one keyword corresponding to the at least one conversation/chat, and controlling the control unit so as to display summary data including the at least one keyword (Han, abstract, [0009]). The artificial intelligence is able to discern when the topic/theme deviates within a chat as depicted in Figures 8A-C, 9 and 10 (Han, [0255]-[0264]). At least Figure 23 depicts selectable generated summaries of distinct chats, that when selected, allows a user to resume the selected chat (Han, [0412]-[0420]). The summary information is updated when a new chat is received (Han, [0421]). The summary information includes a time stamp which is updated when a new chat is received (Han, [0424]-[0425]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Han that a method for generating descriptors of threads would include a user interface displaying generated and selectable up-to-date descriptors. With Hernandez and Han disclosing generating summaries of multiple conversations, and with Han additionally disclosing a user interface that lists selectable summaries of chats, one of ordinary skill in the art of implementing a method for generating descriptors of threads would include a user interface displaying generated and selectable up-to-date descriptors in order to allow a user to resume the conversation of the appropriate chats/threads. One would therefore be motivated to combine these teachings as in doing so would create this method for generating descriptors of threads.
Regarding claim 10, Hernandez and Han teach the computer-implemented method of claim 9, wherein the first selectable UI element further includes a timestamp. More specifically, Hernandez depicts the summaries as including a timestamp (Hernandez, at least Figures 5A, 5B, and 6B-6F). Additionally, Han discloses the summary information includes a time stamp which is updated when a new chat is received (Han, [0424]-[0425]).
Regarding claim 11, Hernandez teaches a computer-implemented method for capturing dynamic states of threads, the method comprising:
generating a first thread descriptor for a first thread at a first state after a turn; generating an updated first thread descriptor for the first thread at a second state after a subsequent turn; storing the updated first thread descriptor with the first thread. More specifically, customers initiate chat sessions with agents; the customers can submit queries, and the agents return responses or vice versa (Hernandez, [0039]-[0042]). Customer input data can be one message and agent input data can be one message (query-response pair) and is associated with a first state. Data (prompt) is generated with the customer input data and agent input data (query-response) and sent to a large language model for generating a summary (descriptor) (Hernandez, [0046]). A first summary (thread descriptor) is generated with the large language model and the agents chat interface is updated with the first summary (Hernandez, abstract, [0046]). At a second state, a second query and second response can be used (Hernandez, abstract, [0065]). The large language model is sent a prompt with the second customer message query and second agent message (second query-response) (Hernandez, abstract, [0065]). The large language model outputs a second summary (second thread descriptor) and updates the agents chat interface with the second summary (Hernandez, abstract, [0065]).
generating a second thread descriptor for a second thread at a first state after a turn; generating an updated second thread descriptor for the second thread at a second state after a subsequent turn; storing the updated second thread descriptor with the second thread. More specifically, Figure 5B depicts a display depicting two conversations and their summaries generated by the process cited above (Hernandez, Figure 5B, [0109]).
However, Hernandez may not explicitly teach every aspect of
generating a user interface comprising: a first selectable user interface (UI) element representing the first thread, the first selectable UI element including the updated first thread descriptor; and a second selectable UI element representing the second thread, the second selectable UI element including the updated second thread descriptor.
Han discloses an artificial intelligence unit for acquiring at least one conversation, acquiring at least one keyword corresponding to the at least one conversation/chat, and controlling the control unit so as to display summary data including the at least one keyword (Han, abstract, [0009]). The artificial intelligence is able to discern when the topic/theme deviates within a chat as depicted in Figures 8A-C, 9 and 10 (Han, [0255]-[0264]). At least Figure 23 depicts selectable generated summaries of distinct chats, that when selected, allows a user to resume the selected chat (Han, [0412]-[0420]). The summary information is updated when a new chat is received (Han, [0421]). The summary information includes a time stamp which is updated when a new chat is received (Han, [0424]-[0425]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Han that a method for generating up-to-date descriptors of threads would include a user interface displaying generated and selectable up-to-date descriptors. With Hernandez and Han disclosing generating summaries of multiple conversations, and with Han additionally disclosing a user interface that lists selectable summaries of chats, one of ordinary skill in the art of implementing a method for generating up-to-date descriptors of threads would include a user interface displaying generated and selectable up-to-date descriptors in order to allow a user to resume the conversation of the appropriate chats/threads. One would therefore be motivated to combine these teachings as in doing so would create this method for generating descriptors of threads.
Regarding claim 12, Hernandez and Han teach the computer-implemented method of claim 11, further comprising: receiving a selection of the first selectable UI element; and based on receiving the selecting, opening the thread in the second state. More specifically, at least Figure 23 depicts selectable generated summaries of distinct chats, that when selected, allows a user to resume the selected chat (Han, [0412]-[0420]).
Claim(s) 13-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez and Han, and further in view of Griesbach et al. (US 2023/0370410 A1, hereinafter “Griesbach”).
Regarding claim 13, Hernandez and Han teach the computer-implemented method of claim 11, including the that the prompt for generating a chat summary includes the previously generated summary, and determining whether a generated summary for a chat/conversation has different information from a previously generated summary, and either updating the summary or creating a new summary for a separate portion of the chat (Hernandez, [0015], [0062], [0069], [0071]).
However, Hernandez and Han may not explicitly teach every aspect of
further comprising: receiving an additional turn to the first thread, thereby forming a third state for the first thread; determining that the additional turn diverges from a prior state of the first thread; based on the additional turn diverging from the first thread, generating a prompt including the updated first thread descriptor and the turn wherein the turn is a query and a response to the query; providing the prompt to a generative AI model; receiving, from the generative AI model in response to the prompt, a new thread descriptor, representing the third state of the first thread; storing the new thread descriptor with the first thread; and surfacing the new thread descriptor, wherein the new thread descriptor replaces the updated first thread descriptor.
Griesbach discloses detecting at least one question (query) in a message thread and to determine if the message thread includes at least one answer (response) responding to the at least one question. From this, the system generates a summary of the message thread based on the detection of the at least one question and the determination of the at least one answer, and outputs for display the summary of the message thread to a recipient of the message thread (Griesbach, abstract). The processor may identify the fork email thread by determining if its content presented therein materially deviates from the original topic. The processor may in turn generate an individual summary for each fork email thread (Griesbach, [0011], [0086]-[0087], Figure 10). The summaries are generated using a machine learning model (Griesbach, [0040], [0060], [0102]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Han with Griesbach that a method for generating descriptors of threads including determining whether a subsequent generated descriptor has different information would include determining whether the thread diverts from the previously generated descriptor and generating a prompt for the generating and displaying the new descriptor according to the divergence. With Hernandez, Han, and Griesbach disclosing generating summaries of conversations, with Hernandez additionally suggesting that a second summary could either be used to update the previous summary or be a new summary for a chat, and with Griesbach additionally suggesting determining when a topic of the thread forks based on a question and answer, and generating a new summary accordingly, one of ordinary skill in the art of implementing a method for generating descriptors of threads including determining whether a subsequent generated descriptor has different information would include determining whether the thread diverts from the previously generated descriptor and generating a prompt for the generating and displaying the new descriptor according to the divergence in order to allow a user to understand the most up-to-date subject matter of the conversation in the appropriate chats/threads. One would therefore be motivated to combine these teachings as in doing so would create this method for generating descriptors of threads.
Regarding claim 14, Hernandez, Han, and Griesbach suggest the computer-implemented method of claim 13, wherein the determining that the additional turn diverges further comprises: generating an embedding for the additional turn; and comparing the embedding for the additional turn to one or more embeddings for previous turns within the first thread. More specifically, this claim is suggested by the combination of Hernandez, Han, and Griesbach where an embedding is generated based on the generated summaries, which in turn can be based on one query-response pair; the embeddings are used to determine where a subsequent summary is different from a prior summary for updating the summary (Hernandez, [0049]-[0050], [0066], [0070] [0120]-[0122]), and determining that a fork in the thread occurs and generating a new summary (Griesbach, [0011], [0086]-[0087], Figure 10).
Regarding claim 15, Hernandez, Han, and Griesbach suggest the computer-implemented method of claim 13, wherein the determining that the additional turn diverges further comprises: generating a divergence prompt including the additional turn and at least one of a prior query, a prior response, or a prior thread descriptor; providing the divergence prompt to the generative AI model; receiving a divergence output from the generative AI model; and determining that the additional turn diverges based on the divergence output. More specifically, this claim is suggested by the combination of Hernandez, Han, and Griesbach where a prompt for a new descriptor includes a previous descriptor as well as a query-response pair (Hernandez, [0015], [0046], [0062], [0069], [0071]), determining that a query-response pair diverges from the previous descriptor (Griesbach, [0011], [0086]-[0087], Figure 10), and determining that a second generated descriptor is different from the prior descriptor (Hernandez, [0015], [0062], [0069], [0071]).
Claim(s) 16 and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez and Han, and further in view of Tanikella.
Regarding claim 16, Hernandez and Han teach the computer-implemented method of claim 11, however, may not explicitly teach every aspect of wherein the first selectable UI element includes at least two of a title, a synopsis, an image, a time stamp, and a thread preview for the first thread.
Tanikella discloses techniques for generating summary documents from conversations within a communication platform. A machine-learned summarization model can receive some or all of the contents of a virtual space and can generate a summarization document reflecting the contents of various communications (Tanikella, abstract). The summary document may include various visualizations, such as but not limited to, a cluster of profile photo of icons representing engagements of users in the conversation, a cluster of emojis used in the conversations representing the sentiments of the conversation, a compilation of visual elements in a tree structure representing the intensity and evolution of user reactions and engagements in the conversation, and/or the like (Tanikella, [0022]). The communication platform can treat the summary document as a synopsis or outline of the virtual space and the posts, events, etc. that took place within the virtual space (Tanikella, [0023]). The summary can consist of a title (Tanikella, [0129]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Han with Tanikella that a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is at least two of a title, a synopsis, an image, a time stamp, and a thread preview for the first thread. With Hernandez and Han with Tanikella disclosing generating summaries of communications, and with Tanikella additionally disclosing the summaries include a title, synopsis, or an visual/image, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts would include that the descriptor is at least two of a title, a synopsis, an image, a time stamp, and a thread preview for the first thread in order to provide a more user friendly way of distinguishing the summaries from each other. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
Regarding claim 17, Hernandez and Han with Tanikella teach the computer-implemented method of claim 16, wherein the first selectable UI element includes at least the title and the image. More specifically, the summary document may include various visualizations, such as but not limited to, a cluster of profile photo of icons representing engagements of users in the conversation, a cluster of emojis used in the conversations representing the sentiments of the conversation, a compilation of visual elements in a tree structure representing the intensity and evolution of user reactions and engagements in the conversation, and/or the like (Tanikella, [0022]). The communication platform can treat the summary document as a synopsis or outline of the virtual space and the posts, events, etc. that took place within the virtual space (Tanikella, [0023]). The summary can consist of a title (Tanikella, [0129]).
Claim(s) 18 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez, and further in view of Perkins et al. (US 2025/0078822 A1, filed 8/28/2023, hereinafter “Perkins”).
Regarding claim 18, Hernandez system for generating a dynamic representation for a thread, comprising: at least one processor; and memory storing instructions that, when executed by the at least one processor, cause the system to perform operations comprising:
receive a first input query for the thread; receive, …, a first response to the first query. More specifically, customers initiate chat sessions with agents; the customers can submit queries, and the agents return responses or vice versa (Hernandez, [0039]-[0042]).
generate a first descriptor prompt including at least one of the first input query and the first response; provide the first descriptor prompt to a … generative AI model. More specifically, customer input data can be one message and agent input data can be one message (query-response pair) and is associated with a first state. Data (prompt) is generated with the customer input data and agent input data (query-response) and sent to a large language model for generating a summary (descriptor) (Hernandez, [0046]).
receive, from the … generative AI model in response to the first descriptor prompt, a first thread descriptor representative of a first state of the thread; surface the first thread descriptor. More specifically, a first summary (thread descriptor) is generated with the large language model and the agents chat interface is updated with the first summary (Hernandez, abstract, [0046]).
receive a second input query for the thread; receive, …, a second response to the second input query. More specifically, at a second state (a second query and second response can be used) (Hernandez, abstract, [0065]).
generate a second descriptor prompt including at least one of the second input query and the second response; provide the second descriptor prompt to the … generative AI model. More specifically, the large language model is sent a prompt with the second customer message query and second agent message (second query-response) (Hernandez, abstract, [0065]).
receive, from the generative AI model in response to the second descriptor prompt, a second thread descriptor representative of a second state of the thread; and surface the second thread descriptor, wherein the second thread descriptor replaces the first thread descriptor. More specifically, the large language model outputs a second summary (second thread descriptor) and updates the agents chat interface with the second summary (Hernandez, abstract, [0065]).
However, Hernandez may not explicitly teach every aspect of
[the first and second response is received] from a first generative artificial intelligence (AI) model;
[the first and second descriptor prompt is provided] to a second generative artificial intelligence (AI) model; [and]
[the first and second thread descriptors is received] from the [second] generative artificial intelligence (AI) model.
Perkins discloses a communications session with a user may be automated using a language model. The language model may be instructed to select a next action to be performed where the next action may include transmitting a responsive communication to the user. Prompts used to query the language model may include a representation of text of the communications session (Perkins, abstract). A user submits a customer support query, and the language model returns a response (Perkins, [0039]). Generating a summary of the communications can be performed by either the same language model used to for the customer support process or a different mathematical model or language model (Perkins, [0062]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Perkins that a method for generating descriptors for threads would include using a first generative model for responding to queries from users in the conversation and a different generative model for generating descriptors of the threads. With Hernandez and Perkins disclosing generating responses to queries from users in conversations as well as generating summaries of conversations, and with Perkins additionally suggesting using a separate generative models for generating responses to queries and generating summaries, one of ordinary skill in the art of implementing a method for generating descriptors for threads would include using a first generative model for responding to queries from users in the conversation and a different generative model for generating descriptors of the threads in order to reduce processing requirements for the generative models. One would therefore be motivated to combine these teachings as in doing so would create this method for generating descriptors for threads.
Regarding claim 20, Hernandez and Perkins teach the system of claim 18, wherein the first thread descriptor includes at least one of a title, a synopsis, or an image. More specifically, the program utilizes the identified information (including an identified topic, construable as a descriptor) as inputs to a trained GAN (i.e., image generator model 156) creating image representations of conversations (Wilson, abstract, [0010], [0023], [0037]).
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hernandez and Perkins, and further in view of Wilson.
Regarding claim 19, Hernandez and Perkins teach the system of claim 18, however, may not explicitly teach every aspect of wherein the operations further comprise: generate a prompt that includes the second thread descriptor; provide the prompt to an image-generating generative AI model; and receive, from the image-generating generative AI model, an image based on at least the second thread descriptor.
Wilson discloses an approach to generating image representations of a conversation between a plurality of users. Generating one or more image representations of the one or more detected utterances utilizes a generative adversarial network restricted by one or more user privacy parameters, wherein the generative adversarial network is fed with an extracted sentiment, a generated avatar, an identified topic, an extracted location, and one or more user preferences. Embodiments of the present invention summarize conversations and display textual topics that are difficult to convey without visual cues. The program identifies users, sentiment, and location in a conversation (e.g., textual or auditory) utilizing a plurality of models (i.e., user model 152, location model 154, etc.).The program utilizes the identified information (including an identified topic, construable as a descriptor) as inputs to a trained GAN (i.e., image generator model 156) creating image representations of conversations (Wilson, abstract, [0010], [0023], [0037], at least user preferences or privacy parameters can be construed as static instructions for generating the image). The conversations are any of email, instant message, direct message, text message, social media post, spoken message, etc. (Wilson, [0026]).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention given the teachings of Hernandez and Perkins with Wilson that a method for generating a descriptor of a thread of communications would include the descriptor including a visual/image generated with a prompt including the descriptor. With Hernandez, Perkins, and Wilson disclosing generating summaries of conversations, with Wilson disclosing the summaries including an visual/image generated with a prompt including an identified topic/summary, one of ordinary skill in the art of implementing a method for generating a descriptor of a thread of communications using generated prompts, the descriptor being one of a title, synopsis, or an image would include that generating an image representation of the descriptor uses a prompt with static instructions for generating the image in order to provide a more user friendly way of distinguishing the summaries from each other. One would therefore be motivated to combine these teachings as in doing so would create this method for generating a descriptor of a thread of communications using generated prompts.
Pertinent Prior Art
The prior art made of record on form PTO-892 and not relied upon is considered pertinent to applicant's disclosure. Applicant is required under 37 C.F.R. § 1.111(c) to consider these references fully when responding to this action.
Dogget (US 2024/0394965 A1) – generates continuously updated summaries of conversations with a virtual character using a machine learning model.
Mauer (US 2024/0176960 A1) – generates continuously updated summaries of conversations using a machine learning model.
Wang (US 2021/0103636 A1) – generating thread summaries with a machine learning model.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PATRICK F RIEGLER whose telephone number is (571)270-3625. The examiner can normally be reached M-F 9:30am-6:00pm, ET.
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, Kieu Vu can be reached at (571) 272-4057. 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.
/PATRICK F RIEGLER/ Primary Examiner, Art Unit 2171