DETAILED ACTION
This communication is in response to the Application filed on January 13, 2025.
Claims 1 - 20 are pending and have been examined.
Claims 1, 13 and 19 are independent.
Foreign priority: July 12, 2024.
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 statements (IDS) submitted on October 16, 2024, March 11, 2026 and July 9, 2026 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements were considered by the examiner.
Specification
Applicant is reminded of the proper language and format for an abstract of the disclosure.
The abstract should be in narrative form and generally limited to a single paragraph on a separate sheet within the range of 50 to 150 words. The form and legal phraseology often used in patent claims, such as "means" and "said," should be avoided. The abstract should describe the disclosure sufficiently to assist readers in deciding whether there is a need for consulting the full patent text for details.
The abstract of the disclosure is objected to because the abstract of 167 words exceeds 150 words. Correction is required. See MPEP § 608.01(b).
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.
Claim 20 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Regarding Claim 20,
While Claim 20 appears to be a dependent claim, it is rather an independent claim where “the steps of the method of claim 1 are implemented” only and not further limiting. Claim 20 is clearly a program (“A computer program product comprising computer programs…”).
Step 1: Claim 20 recites “A computer program product comprising computer programs…” for role-playing of a large model enabling customizable, anthropomorphic chat robot. “Computer program” is not directed to any statutory category. The claim is not patent eligible. See MPEP 2106.03. (Step 1: No)
Claim Rejections - 35 USC § 103
The following is a quotation of pre-AIA 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
Claims 1 - 5, 7, 9 - 17 and 19 - 20 are rejected under 35 U.S.C. 103(a) as being unpatentable over Ju et al., (CN118403367A), hereinafter referred to as Ju, in view of Chen, (CN116680391A).
Regarding Claims 1, 13 and 19, Ju teaches:
1. A method for training a large model, comprising, 13. An electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the processor is configured to, and 19. A non-transitory computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are used to cause a computer to perform the method for training a large model, comprising:
obtaining a conversation sample, wherein the conversation sample comprises portraits of a plurality of roles, [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”);” Par. n0008]
a plot containing the plurality of roles and a plurality of rounds of conversations among the plurality of roles; [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”);” Par. n0008; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079]
for any one of the plurality of roles, obtaining a predicted conversation sentence of the role by inputting the portraits of the plurality of roles, [Ju teaches language model can be BERT (Par. n0051) which is well known to those skilled in the art for next sentence prediction: “The initial language model can be an open-source language model, including but not limited to: Mistral, OpenAI GPT, BERT, etc.” Par. n0051; “using the first language model and the second language model (Ju teaches language model can be BERT (Par. n0051) which is well known to those skilled in the art for next sentence prediction) to alternately output statements (i.e., the claimed “predicted conversation sentence”) to obtain the dialogue data (i.e., the claimed “predicted conversation sentence”) between the first persona and the second persona.” Par. n0009]
the plot and a historical conversation sentence corresponding to a sample conversation sentence of the role in the plurality of rounds of conversations into an initial large model; and [Ju, Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “plot and historical conversation sentence corresponding to a sample conversation sentence”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sentence”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”.]
obtaining a target large model by training the initial large model according to a difference between the predicted conversation sentence and the sample conversation sentence. [Ju, “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data includes one or more rounds of dialogue data between different characters;” Par. n0012; “loss value (i.e., the claimed “difference”) between the character description dataset output by the initial language model (i.e., the claimed “predicted conversation sentence”) and the character sample features (i.e., the claimed “sample conversation sentence”); iteratively updating the model parameters of the initial language model based on the loss value (i.e., the claimed “difference”) to obtain the character generation model (i.e., the claimed “target large model”).” Par. n0005]
Ju fails to explicitly teach predict.
However, Chen teaches:
for any one of the plurality of roles, obtaining a predicted conversation sentence of the role by inputting the portraits of the plurality of roles, [Chen, “dialogue data (i.e., the claimed “predicted conversation sentence”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; “dialogue statement are predicted to obtain the response statement (i.e., the claimed “predicted conversation sentence”) that conforms to the role setting of the first role.” Par. n0013; “loss function based on the sample dialogue statements and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304]
obtaining a target large model by training the initial large model according to a difference between the predicted conversation sentence and the sample conversation sentence. [Chen, “loss (i.e., the claimed “difference”) function based on the sample dialogue statements (i.e., the claimed “sample conversation sentence”) and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304]
Ju and Chen pertain to role playing generation systems and are analogous to the instant application. Accordingly, it would have been obvious to one of ordinary skill in the role playing generation systems art to modify Ju’s teachings of “multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”)” (Ju, Par. n0008) with the explicit teachings of “predicted response statements (i.e., the claimed “predicted conversation sentence”)” (Chen, Par. n0304) taught by Chen in order to “quickly realize customized dialogue for multiple roles” (Chen, Par. n0004).
Regarding Claims 2 and 14, Ju in view of Chen has been discussed above. The combination further teaches:
obtaining the portraits of the plurality of roles and the plot; [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”);” Par. n0008; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; Chen, “dialogue data of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
for any one of the plurality of roles, during a conversation process of the plurality of roles, obtaining a sample conversation sentence of the role according to the portraits of the plurality of roles and the plot; [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “sample conversation sentence”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”);” Par. n0008; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; Chen, “dialogue data (i.e., the claimed “sample conversation sentence”) of different roles (i.e., the claimed “of the role according to the portraits of the plurality of roles”),” Par. n0003]
obtaining the plurality of rounds of conversations among the plurality of roles according to sample conversation sentences of the plurality of roles in the conversation process of the plurality of roles; and [Ju, “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “plot and historical conversation sentence corresponding to a sample conversation sentence”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sentence”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”; Chen, “dialogue data of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003
obtaining the conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations. [Ju, “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
Regarding Claims 3 and 15, Ju in view of Chen has been discussed above. The combination further teaches:
obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles and the plot by calling a first large model corresponding to the role, wherein different roles of the plurality of roles correspond to different first large models. [Ju, “using the first language model and the second language model (i.e., the claimed “different first large model”) to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
Regarding Claims 4 and 16, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles and the plot by calling the first large model corresponding to the role, comprises: [Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
obtaining a language style of the role; and [Ju, “character traits (i.e. the claimed “language style”), Par. n0027; “language expressions that conform to the character setting’s (i.e., the claimed “language style”) gender, personality (i.e., the claimed “language style”, character traits (i.e., the claimed “language style”), occupation, and other characteristics.” Par. n0002; Chen, “adding a character model part corresponding to the first character to a general dialogue model, and the character model part is used to enable the customized dialogue model to generate response statements that conform to the character settings of the first character (i.e., the claimed “language style of the role”).” Par. n0008; “Optionally, character settings (i.e., the claimed “language style of the role”) include at least one of the following: character appearance, clothing style, physical characteristics, personality traits (i.e., the claimed “language style of the role”), life background, knowledge background, plot background, scene background, and catchphrase (i.e., the claimed “language style”).” Par. n0079; Referring to the Specification of the instant Application, “language style” refers to “speaking style of the role”.]
obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles, [Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
the plot and the language style by calling the first large model corresponding to the role. [Ju, “language expressions that conform to the character setting’s (i.e., the claimed “language style”) gender, personality (i.e., the claimed “language style”, character traits (i.e., the claimed “language style”), occupation, and other characteristics.” Par. n0002; Chen, “adding a character model part corresponding to the first character to a general dialogue model, and the character model part is used to enable the customized dialogue model to generate response statements that conform to the character settings of the first character (i.e., the claimed “language style of the role”).” Par. n0008; “Optionally, character settings (i.e., the claimed “language style of the role”) include at least one of the following: character appearance, clothing style, physical characteristics, personality traits (i.e., the claimed “language style of the role”), life background, knowledge background, plot background, scene background, and catchphrase (i.e., the claimed “language style”).” Par. n0079; Referring to the Specification of the instant Application, “language style” refers to “speaking style of the role”; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
Regarding Claims 5 and 17, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles and the plot, comprises: [Ju, Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
determining a target round whose conversation difficulty is to be increased and a target strategy for increasing the conversation difficulty corresponding to the target round; [Ju, “You can alternately input the existing dialogue context and character description datasets into the corresponding character language model to obtain the corresponding character’s statements, thus obtaining more rounds of dialogue data (i.e., larger amount of dialogue data is the claimed “increasing conversation difficulty corresponding to the target round”).” Par. n0077; Referring to the Specification of the instant Application “conversation difficulty” refers to “amount of information”/ “larger the amount of information”.]
in a conversation of the target round, obtaining a candidate conversation sentence of the role according to the portraits of the roles and the plot; and [Ju, Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment (i.e., the claimed “conversation of the target round”) of the multiple character dialogue data (i.e., the claimed “conversation sentence of the role according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “candidate conversation sentence”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”; Chen, “dialogue data (i.e., the claimed “candidate conversation sentence”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
obtaining the sample conversation sentence of the role by adopting the target strategy according to the portraits of the roles, the plot, the candidate conversation sentence and a historical conversation sentence of the candidate conversation sentence. [Ju, “You can alternately input the existing dialogue (i.e., the claimed “historical conversation sentence”) context and character description datasets into the corresponding character language model to obtain the corresponding character’s statements, thus obtaining more rounds of dialogue data (i.e., larger amount of dialogue data is the claimed “increasing conversation difficulty corresponding to the target round”).” Par. n0077; Referring to the Specification of the instant Application “conversation difficulty” refers to “amount of information”/ “larger the amount of information”; Ju, Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; each segment (i.e., the claimed “conversation of the target round”) of the multiple character dialogue data (i.e., the claimed “conversation sentence of the role according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “candidate conversation sentence”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”]
Regarding Claim 7, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the portraits of the plurality of roles and the plot, comprises: obtaining the portraits of the plurality of roles; and [Ju, Ju, “selecting different characters from different character categories (i.e., the claimed “obtaining the portraits of the plurality of roles”) and generating dialogue data between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
generating the plot according to the portraits of the plurality of roles by calling a second large model. [Ju, Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “generate the plot”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
Regarding Claim 9, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the portraits of the plurality of roles and the plot, comprises: [Ju, “selecting different characters from different character categories (i.e., the claimed “obtaining the portraits of the plurality of roles”) and generating dialogue data between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
obtaining a portrait of a target role of the plurality of roles; [Ju, “selecting different characters (i.e., the claimed “obtaining a portrait of a target role”) from different character categories (i.e., the claimed “plurality of roles”) and generating dialogue data between the different characters,” Par. n0011; Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “obtaining the plot”) between the first persona and the second persona (i.e., the claimed “according to the portrait of the target role”).” Par. n0009]
obtaining the plot according to the portrait of the target role by calling a second large model; and [Ju, “selecting different characters (i.e., the claimed “according to the portrait of the target role”) from different character categories and generating dialogue data (i.e., the claimed “obtaining the plot”) between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “obtaining the plot”) between the first persona and the second persona (i.e., the claimed “according to the portrait of the target role”).” Par. n0009]
obtaining portraits of other roles of the plurality of roles except the target role according to the plot by calling a third large model. [Ju, Ju, “selecting different characters from different character categories (i.e., the claimed “obtaining portraits of other roles of the plurality of roles except the target role”) and generating dialogue data between the different characters,” Par. n0011; “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; Claim is directed to repeating the subject matter for obtaining more portraits and another large model. However, obtaining more portraits / another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale.]
Regarding Claim 10, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the portraits of the plurality of roles and the plot, comprises: [Ju, “selecting different characters from different character categories (i.e., the claimed “obtaining the portraits of the plurality of roles”) and generating dialogue data between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”; Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
obtaining the plot by calling a second large model; and [Ju, “selecting different characters (i.e., the claimed “according to the portrait of the target role”) from different character categories and generating dialogue data (i.e., the claimed “obtaining the plot”) between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “obtaining the plot”) between the first persona and the second persona (i.e., the claimed “according to the portrait of the target role”).” Par. n0009]
obtaining portraits of a plurality of roles in the plot according to the plot by calling a third large model. [Ju, “selecting different characters (i.e., the claimed “according to the portrait of the target role”) from different character categories and generating dialogue data (i.e., the claimed “obtaining the plot”) between the different characters,” Par. n0011; “Optionally, character settings include at least one of the following: character appearance, clothing style, physical characteristics, personality traits, life background, knowledge background, plot background, scene background, and catchphrase.” Par. n0079; “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data (i.e., the claimed “conversation sample according to the portraits of the plurality of roles, the plot and the plurality of rounds of conversations”) includes one or more rounds (i.e., the claimed “plurality of rounds”) of dialogue (i.e., the claimed “conversations”) data (i.e., the claimed “conversation sample”) between different characters (i.e., the claimed “roles”);” Par. n0012; Referring to Specification of the instant Application Pg. 1: “plot” contains “plurality of roles and a plurality of rounds of conversations among the plurality of roles”;’ Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “obtaining the plot”) between the first persona and the second persona (i.e., the claimed “according to the portrait of the target role”).” Par. n0009; “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; Claim is directed to repeating the subject matter for obtaining more portraits and another large model. However, obtaining more portraits / another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale.]
Regarding Claim 11, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining a conversation sample, comprises: obtaining a plurality of conversation samples; [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”);” Par. n0008; Chen, “dialogue data (i.e., the claimed “conversation sample”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003]
wherein obtaining the target large model by training the initial large model according to the difference between the predicted conversation sentence and the sample conversation sentence, comprises: [Ju, “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data includes one or more rounds of dialogue data between different characters;” Par. n0012; “loss value (i.e., the claimed “difference”) between the character description dataset output by the initial language model (i.e., the claimed “predicted conversation sentence”) and the character sample features (i.e., the claimed “sample conversation sentence”); iteratively updating the model parameters of the initial language model based on the loss value (i.e., the claimed “difference”) to obtain the character generation model (i.e., the claimed “target large model”).” Par. n0005; Chen, “dialogue data (i.e., the claimed “predicted conversation sentence”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; “dialogue statement are predicted to obtain the response statement (i.e., the claimed “predicted conversation sentence”) that conforms to the role setting of the first role.” Par. n0013; “loss function based on the sample dialogue statements and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304; Chen, “loss (i.e., the claimed “difference”) function based on the sample dialogue statements (i.e., the claimed “sample conversation sentence”) and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304]
for a current conversation sample, obtaining a first intermediate large model by training the initial large model according to the difference between the predicted conversation sentence and the sample conversation sentence; [Ju, “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data includes one or more rounds of dialogue data between different characters;” Par. n0012; “loss value (i.e., the claimed “difference”) between the character description dataset output by the initial language model (i.e., the claimed “predicted conversation sentence”) and the character sample features (i.e., the claimed “sample conversation sentence”); iteratively updating the model parameters of the initial language model based on the loss value (i.e., the claimed “difference”) to obtain the character generation model (i.e., the claimed “target large model”).” Par. n0005; Chen, “dialogue data (i.e., the claimed “predicted conversation sentence”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; “dialogue statement are predicted to obtain the response statement (i.e., the claimed “predicted conversation sentence”) that conforms to the role setting of the first role.” Par. n0013; “loss function based on the sample dialogue statements and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304; Chen, “loss (i.e., the claimed “difference”) function based on the sample dialogue statements (i.e., the claimed “sample conversation sentence”) and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304; “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; Claim is directed to repeating the subject matter for another large model. However, another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale.]
obtaining a second intermediate large model by training the first intermediate large model by using a next conversation sample, [Ju, “the character generation model (i.e., the claimed “target large model”) is obtained by training an initial language model using character sample features and character sample dialogue data; each segment of the multiple character dialogue data includes one or more rounds of dialogue data between different characters;” Par. n0012; “loss value (i.e., the claimed “difference”) between the character description dataset output by the initial language model (i.e., the claimed “predicted conversation sentence”) and the character sample features (i.e., the claimed “sample conversation sentence”); iteratively updating the model parameters of the initial language model based on the loss value (i.e., the claimed “difference”) to obtain the character generation model (i.e., the claimed “target large model”).” Par. n0005; Chen, “dialogue data (i.e., the claimed “predicted conversation sentence”) of different roles (i.e., the claimed “portraits of the plurality of roles”),” Par. n0003; “dialogue statement are predicted to obtain the response statement (i.e., the claimed “predicted conversation sentence”) that conforms to the role setting of the first role.” Par. n0013; “loss function based on the sample dialogue statements and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304; Chen, “loss (i.e., the claimed “difference”) function based on the sample dialogue statements (i.e., the claimed “sample conversation sentence”) and the predicted response statements (i.e., the claimed “predicted conversation sentence”).” Par. n0304; “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; Claim is directed to repeating the subject matter for another large model. However, another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale.]
wherein a difficulty level of the current conversation sample is less than a difficulty level of the next conversation sample; and [Ju clearly teaches obtaining more rounds of dialogue data increases difficulty. It is obvious to one skilled in the art that the existing/prior dialogue round has less dialogue data and therefore less difficulty: “You can alternately input the existing dialogue context and character description datasets into the corresponding character language model to obtain the corresponding character’s statements, thus obtaining more rounds of dialogue data (i.e., larger amount of dialogue data is the claimed “increasing conversation difficulty corresponding to the target round”).” Par. n0077; Referring to the Specification of the instant Application “conversation difficulty” refers to “amount of information”/ “larger the amount of information”.
continuing to train the second intermediate large model with a next conversation sample of the next conversation sample until the target large model is obtained, [ “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; “train the initial role-playing model (i.e., the claimed “second intermediate large model”) to obtain a role-playing model (i.e., the claimed “target large model”).” Par. n0011; Claim is directed to repeating the subject matter for another large model. However, another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale]
wherein a difficulty level of the next conversation sample is less than the difficulty level of the previous next conversation sample. [Ju, see mapping applied to claims 1-5; Chen, see mapping applied to claims 1-5; Ju clearly teaches obtaining more rounds of dialogue data increases difficulty. It is obvious to one skilled in the art that obtaining more rounds of dialogue data increase difficulty with each round: “You can alternately input the existing dialogue context and character description datasets into the corresponding character language model to obtain the corresponding character’s statements, thus obtaining more rounds of dialogue data (i.e., larger amount of dialogue data is the claimed “increasing conversation difficulty corresponding to the target round”).” Par. n0077; Referring to the Specification of the instant Application “conversation difficulty” refers to “amount of information”/ “larger the amount of information”.]
Regarding Claim 12, Ju in view of Chen has been discussed above. The combination further teaches:
obtaining a new conversation sample based on the conversation sample, [Ju, “Specifically, the first segment of character dialogue data can be the dialogue data (i.e., the claimed “conversation sample”) between tour guide A and tourist B, and the second segment of character dialogue data (i.e., the claimed “new conversation sample”) can be the group chat dialogue data between tour guide A and tourists C and D.” Par. n0036]
wherein the sample conversation sentence of any role in the new conversation sample is obtained by calling the target large model; and [Ju, “Specifically, the first segment of character dialogue data can be the dialogue data (i.e., the claimed “conversation sample”) between tour guide A and tourist B, and the second segment of character dialogue data (i.e., the claimed “new conversation sample”) can be the group chat dialogue data between tour guide A and tourists C and D.” Par. n0036; Specifically, two or more language models (i.e., the claimed “target large model”) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044]
continuing to train the target large model using the new conversation sample. [“train the initial role-playing model (i.e., the claimed “second intermediate large model”) to obtain a role-playing model (i.e., the claimed “target large model”).” Par. n0011]
Regarding Claim 20, Ju in view of Chen has been discussed above. The combination further teaches:
wherein when the computer programs are executed by a processor, the steps of the method of claim 1 are implemented. [Ju; “One or more embodiments of this specification also provide a role-playing dialogue data generation apparatus, including a storage medium and a processor.” Par. n0003; “One or more embodiments of this specification also provide a computer-readable storage medium storing computer instructions (i.e., the claimed “computer programs”) that, when executed by a processor, enable the role-playing dialogue data generation method,” Par. n0014]
Claims 6 and 18 are rejected under 35 U.S.C. 103(a) as being unpatentable over Ju in view of Chen as applied in claim 2 above, and in further view of Liu et al., (U.S. Patent Application Publication 2024/0354158), hereinafter referred to as Liu.
Regarding Claims 6 and 18, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles and the plot, comprises: [Ju, “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
obtaining the sample conversation sentence of the role according to the portraits of the plurality of roles and the plot. [Ju, “selecting different characters (i.e., the claimed “according to the portraits of the plurality of roles”) from different character categories and generating dialogue data (i.e., the claimed “obtaining the sample conversation sentence”) between the different characters,” Par. n0011; “using the first language model and the second language model to alternately output statements to obtain the dialogue data (i.e., the claimed “sample conversation sentence of the role according to the portraits of the plurality of roles”) between the first persona and the second persona (i.e., the claimed “according to the portraits of the plurality of roles”).” Par. n0009]
The combination fails to explicitly teach verifying the plot.
However, Liu teaches:
verifying the plot; and [Liu, “plot task of obtaining a specified compensation by completing a work plot, and the like, and the task verification (i.e., the claimed “verifying the plot”) operation is triggered in response to the first account completing the job plot task.” Par. 0129]
in response to the plot passing the verification, [Liu, “plot task of obtaining a specified compensation by completing a work plot, and the like, and the task verification (i.e., the claimed “verifying the plot”) operation is triggered in response to the first account completing the job plot task.” Par. 0129]
Ju, Chen and Liu pertain to role playing systems and are analogous to the instant application. Accordingly, it would have been obvious to one of ordinary skill in the role playing systems art to modify Ju’s teachings of “multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”)” (Ju, Par. n0008) with the explicit teachings of “predicted response statements (i.e., the claimed “predicted conversation sentence”)” (Chen, Par. n0304) taught by Chen and plot “task verification (i.e., the claimed “verifying the plot”)” (Liu, Par. 0129) taught by Liu in order to “quickly realize customized dialogue for multiple roles” (Chen, Par. n0004) and improve “multiplayer online role-playing games” (Liu, Par. 0003).
Claim 8 is rejected under 35 U.S.C. 103(a) as being unpatentable over Ju in view of Chen as applied in claim 7 above, and in further view of Hu, (CN116637375A).
Regarding Claim 8, Ju in view of Chen has been discussed above. The combination further teaches:
wherein obtaining the portraits of the plurality of roles comprises at least one of: [Ju, “selecting different characters from different character categories (i.e., the claimed “obtaining the portraits of the plurality of roles”) and generating dialogue data (i.e., the claimed “obtaining the sample conversation sentence”) between the different characters,” Par. n0011]
determining the portraits of the plurality of roles from portraits of a plurality of real roles; [Ju, “selecting different characters (i.e., the claimed “determining the portraits of the plurality of roles”) from different character categories (i.e., the claimed “portraits of a plurality of real roles”) and generating dialogue data (i.e., the claimed “obtaining the sample conversation sentence”) between the different characters,” Par. n0011; Ju “A persona can be a real person (i.e., the claimed “real role”), a virtual persona or an anthropomorphic object, such as a character or an anthropomorphic object in a movie or TV series or a game, or a persona of a professional in a certain industry, or a persona created by the user.” Par. n0026; Ju, “For example, role-playing models can mimic the language interaction between tour guides and tourists (i.e., the claimed “portraits of a plurality of real roles”), providing tour guide-related services to tourists.” Par. n0026]
determining a target portrait attribute from a candidate portrait attribute, [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character (i.e., the claimed “target portrait”) and multiple other characters (i.e., the claimed “candidate portrait”);” Par. n0008]
determining a target attribute value of a target portrait attribute from a candidate attribute value of the target portrait attribute, and [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character (i.e., the claimed “target portrait”) and multiple other characters (i.e., the claimed “candidate portrait”);” Par. n0008]
determining a portrait of the role according to the target attribute value of the target portrait attribute; and [Ju, “According to some embodiments of the role-playing dialogue data generation method described in this specification, the multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character (i.e., the claimed “target portrait”) and multiple other characters (i.e., the claimed “candidate portrait”);” Par. n0008]
obtaining reference portraits, and obtaining the portraits of the plurality of roles according to the reference portraits by calling a third large model. [Ju, “selecting different characters (i.e., the claimed “obtaining reference portrait”) from different character categories (i.e., the claimed “portraits of a plurality of roles”) and generating dialogue data (i.e., the claimed “obtaining the sample conversation sentence”) between the different characters,” Par. n0011, Ju, “Specifically, two or more language models (i.e., the claimed “first large model”, “second large model”, “third large model”, “intermediate large model”, etc.) can be used to generate dialogue sentences based on the selected persona description dataset, thereby obtaining dialogue data between different personas.” Par. n0044; Claim is directed to repeating the subject matter for another portrait and another large model. However, another portrait / another large model / repeating steps known from prior art is straightforward, amounts to the normal use of the teachings of Ju in view of Chen and are rejected under similar rationale.]
The combination fails to explicitly teach attribute.
However, Hu teaches:
determining a target portrait attribute from a candidate portrait attribute, [Hu, “Identify at least one virtual character (i.e., the claimed “determining a target portrait attribute”) to participate in each round of dialogue, and configure the attributes (i.e., the claimed “target portrain attribute”) of the virtual character based on the character’s personality, behavior, relationship, task or clues;” Par. n0020; “For example, first set up a database that includes attribute information (i.e., the claimed “candidate portrait attributes”) such as characters, behaviors, relationships, tasks, and clues.” Par. n0040]
determining a target attribute value of a target portrait attribute from a candidate attribute value of the target portrait attribute, and [Hu, “Identify at least one virtual character (i.e., the claimed “determining a target portrait attribute”) to participate in each round of dialogue, and configure the attributes (i.e., the claimed “target portrain attribute”) of the virtual character based on the character’s personality, behavior, relationship, task or clues;” Par. n0020; “For example, first set up a database that includes attribute information (i.e., the claimed “candidate portrait attributes”) such as characters, behaviors, relationships, tasks, and clues.” Par. n0040; “Task attributes mainly include preset task name (i.e., the claimed “attribute value”) and/or clue description text, activation conditions, prerequisite clues or tasks, and are also affected by identity, personality, behavior, and relationship,” Par. n0044]
determining a portrait of the role according to the target attribute value of the target portrait attribute; and [Hu, “Identify at least one virtual character (i.e., the claimed “determining a target portrait attribute”) to participate in each round of dialogue, and configure the attributes (i.e., the claimed “target portrain attribute”) of the virtual character based on the character’s personality, behavior, relationship, task or clues;” Par. n0020; “For example, first set up a database that includes attribute information (i.e., the claimed “candidate portrait attributes”) such as characters, behaviors, relationships, tasks, and clues.” Par. n0040; “Task attributes mainly include preset task name (i.e., the claimed “attribute value”) and/or clue description text, activation conditions, prerequisite clues or tasks, and are also affected by identity, personality, behavior, and relationship,” Par. n0044]
Ju, Chen and Hu pertain to role playing generation systems and are analogous to the instant application. Accordingly, it would have been obvious to one of ordinary skill in the role playing generation systems art to modify Ju’s teachings of “multiple character (i.e., the claimed “portraits of a plurality of roles”) dialogue data (i.e., the claimed “conversation sample”) includes one or more rounds (i.e., the claimed “plurality of rounds of conversations”) of dialogue data (i.e., the claimed “conversation sample”) between the target character and multiple other characters (i.e., the claimed “portraits of a plurality of roles”)” (Ju, Par. n0008) with the explicit teachings of “predicted response statements (i.e., the claimed “predicted conversation sentence”)” (Chen, Par. n0304) taught by Chen and “attributes” (Hu, Par. n0020) taught by Hu in order to “quickly realize customized dialogue for multiple roles” (Chen, Par. n0004) and enable “real time acquisition of the user’s dialogue intent “ (Hu, Par. n0009).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Graff et al., (U.S. Patent Application Publication 2025/0182764) teaches conversational roles.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EUNICE LEE whose telephone number is 571-272-1886. The examiner can normally be reached M-F 8:00 AM - 5:00 PM.
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, Bhavesh Mehta can be reached on 571-272-7453. 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.
/EUNICE LEE/Examiner, Art Unit 2656
/EDGAR X GUERRA-ERAZO/ Primary Examiner, Art Unit 2656
Prosecution Insights