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 .
Claim Objections
Claims 5, 12, 19 are objected to because of the following informalities:
In Claim 5, line 4, the second “of” was probably meant to be or. The same objection is made for Claims 12 and 19.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Zheng, “BIM-GPT: a Prompt-Based Virtual Assistant Framework for BIM Information Retrieval”, April 2023, in view of Cluff, US 2023/0341998 A1.
Regarding Claim 1, Zheng teaches:
A method, comprising: receiving, by one or more processors via an artificial intelligence (AI) chat interface, a prompt from a user (section 2, p. 8: “This paper presents BIM-GPT, a prompt-based Virtual Assistant (VA) framework for BIM Information Retrieval (IR), designed to overcome limitations of existing approaches found in the literature. The BIM-GPT framework offers a NL interface, coupled with BIM visualization, to facilitate efficient IR and interaction with 3D building components for construction practitioners”);
dynamically determining, by the one or more processors using an AI model, a format and a content of an output user interface, wherein both the format and the content are determined dynamically by the AI model based on the prompt (section 2, p. 8: “The UI module enables users to engage with the VA by entering NL queries and receiving NL responses through chat boxes while visualizing the retrieved results in a cloud-based BIM. Text queries are relayed from the UI module to the NLP module, where responses are subsequently generated. To interpret the user's query, the NLP module dynamically produces prompts and acquires generated texts from an external GPT server. Within the DM module, a cloud-based database hosts the building information and provides results in response to structured queries from the NLP module”. See also Figs. 14 and 15 that shows the dynamically determined format and content by the BIM-GPT in the UI).
With Zheng teaching those parts of the limitation as previously pointed out, Zheng may not have explicitly taught all of the following, however, Cluff shows:
and generating, by the one or more processors using the AI model, the output user interface in the determined format and including the determined content (Abstract: “automatically classify, segment, and parse content data using artificial intelligence and natural language processing technology, and generate graphical user interfaces”). (Emphasis added).
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to use the teachings of Cluff with that of Zheng for generating using the AI model, the output user interface.
The ordinary artisan would have been motivated to modify Zheng in the manner set forth above for the purposes of enabling end users to effectively filter data by choosing between layouts that display one or more of various categories of data [Cluff: Abstract].
Regarding Claim 2, Zheng further teaches:
The method of claim 1, wherein dynamically determining the format of the output user interface comprises selecting, based on the prompt, by the one or more processors using the AI model, the format from among a plurality of predetermined output content formats based on the prompt (subsection 5.1.3, p. 26: “We developed a web-based UI for the VA prototype to support efficient BIM IR. To enable NL interaction, we integrated with the open-source Genie Server [60] developed by Stanford Open Virtual Assistant Lab (OVAL) as the front-end chat interface. The Genie Server was further integrated with Autodesk Forge, which provided a library for visualizing the cloud-based BIM. To enable seamless communication between the front-end interface and the prompt manager, we used web sockets to send and receive NL queries and answers, as well as the IDs of retrieved results. Based on these IDs, the cloud-based BIM rendered different 3D contextual scenes of the retrieved building objects in response to the user's query”. The different 3D contextual scenes of the retrieved building objects in response to the user's query representative of the plurality of predetermined output content formats based on the prompt).
Regarding Claim 3, Zheng further teaches:
The method of claim 1, wherein dynamically determining the format and the content of the output user interface comprises: predicting, by the one or more processors using the AI model, an intent of the user from the prompt (subsection 3.1.1, p. 10: “The effectiveness of the NLP module in the BIM-GPT framework relies heavily on the quality of prompts. To better align GPT models with BIM IR contexts, we designed the dynamic prompt template based on consideration of characteristics of GPT models. Because the prompt serves as the conditional probability for GPT models to predict the next word in a sequential order, the components of the template are arranged in a specific order, with the more pertinent content positioned closer to the end of the prompt. As such, the text towards the end of the prompt tends to carry greater weight than the text at the beginning, which helps GPT models generate relevant and coherent responses”; And, p. 11: “we decompose it into a chain of subproblems, including text classification and information extraction and prediction, which can be handled by the models with appropriate prompting. Specifically, the GPT model first classifies the user intent and the category of building objects with the Intent Prompt, then identifies the parameters of the categorized building objects with the Parameter Prompt, and lastly recognizes the values of the identified parameters with the Value”);
and determining, by the one or more processors using the AI model, the format and the content of the output user interface using the predicted intent (section 2, p. 8: “The UI module enables users to engage with the VA by entering NL queries and receiving NL responses through chat boxes while visualizing the retrieved results in a cloud-based BIM. Text queries are relayed from the UI module to the NLP module, where responses are subsequently generated. To interpret the user's query, the NLP module dynamically produces prompts and acquires generated texts from an external GPT server. Within the DM module, a cloud-based database hosts the building information and provides results in response to structured queries from the NLP module”. See also Figs. 14 and 15 that shows the dynamically determined format and content by the BIM-GPT in the UI).
Regarding Claim 4, Zheng further teaches:
The method of claim 1, wherein the output user interface comprises a predetermined interface with a plurality of populatable interface portions, and wherein generating the output user interface comprises populating, by the AI model, at least one of the one or more populatable interface portions using the dynamically determined format and content (subsection 5.1.3, p. 26: “We developed a web-based UI for the VA prototype to support efficient BIM IR. To enable NL interaction, we integrated with the open-source Genie Server [60] developed by Stanford Open Virtual Assistant Lab (OVAL) as the front-end chat interface. The Genie Server was further integrated with Autodesk Forge, which provided a library for visualizing the cloud-based BIM. To enable seamless communication between the front-end interface and the prompt manager, we used web sockets to send and receive NL queries and answers, as well as the IDs of retrieved results. Based on these IDs, the cloud-based BIM rendered different 3D contextual scenes of the retrieved building objects in response to the user's query”. See also Figs. 14 and 15 that shows the one or more populatable interface portions. See also, Cluff, for example paragraph 5).
Regarding Claim 5, Zheng further teaches:
The method of claim 1, wherein the AI model is a generative AI model and dynamically determining the format and the content comprises dynamically generating, using the generative AI model, at least a portion of the format or the content of the output user interface using the generative AI model based on the prompt, wherein the at least a portion of the format of the content of the output user interface comprises a new format or a new content generated by the generative AI model that was not preexisting, prior to receipt of the prompt from the user, in underlying data upon which the generative AI model was trained (Abstract: “We introduce BIM-GPT, a prompt-based virtual assistant (VA) framework integrating BIM and generative pre-trained transformer (GPT) technologies to support NL-based IR. A prompt manager and dynamic template generate prompts for GPT models, enabling interpretation of users' NL queries, summarization of retrieved information, and answering BIM-related questions”. And, section 2, p. 8: “The UI module enables users to engage with the VA by entering NL queries and receiving NL responses through chat boxes while visualizing the retrieved results in a cloud-based BIM. Text queries are relayed from the UI module to the NLP module, where responses are subsequently generated. To interpret the user's query, the NLP module dynamically produces prompts and acquires generated texts from an external GPT server. Within the DM module, a cloud-based database hosts the building information and provides results in response to structured queries from the NLP module”. See also Figs. 14 and 15 that shows the dynamically determined format and content by the BIM-GPT in the UI. And, subsection 4.3, p. 22: “our prompt-based approach successfully classified the intent and building object category, identified filter and projection parameters, and extracted and predicted parameter values from NL queries. Our dynamic template was successfully employed to generate the Intent, Parameter, and Value Prompts, and achieved high accuracy rates, even in the zero-shot scenario”, the predicted parameter values representative of new content generated by the generative AI model that was not preexisting, prior to receipt of the prompt from the user).
Regarding Claim 6, Zheng further teaches:
The method of claim 1, wherein the output user interface comprises a history of one or more prompts previously received by the AI chat interface (Figs. 14 and 15; subsection 5.2, p. 26: “Fig. 14 displays the developed VA prototype by implementing the BIM-GPT framework. Within the user interface, users input NL queries (shown in green boxes) and receive VA’s NL answers (shown in white boxes) on the left in the chat boxes, while they visualize retrieved results and navigate the 3D model using the GUI of Autodesk Forge on the right”. The user queries and received answers in the left boxes of Figs. 14 and 15 representative of the history of one or more prompts previously received by the AI chat interface).
Regarding Claim 7, Zheng further teaches:
The method of claim 6, wherein the history of one or more prompts is dynamically generated responsive to the prompt in response to the AI model determining the one or more prompts to be potentially relevant to the prompt (Fig. 15; subsection 5.2, p. 27: “In Fig. 15, a dialogue between the user and the VA is presented. Upon initializing the prototype, the BIM of the six-story hospital building was rendered. In this example, the user queried the building components by filtering on the level_number, component_type and component_id, and the corresponding retrieved results were visualized in the cloud based BIM as shown in (a), (b), and (c), respectively. The summarized results, along with their 3D contexts, helped the user retrieve the required information and easily navigate the BIM. Finally, the user posed a general question about the building system attribute of the pump, and the VA’s answer provided additional information, facilitating the IR process”. As seen in the figure, all prompt histories as shown are potentially relevant (relates to the same topic/pump) to the final prompt).
Claims 8-14 are similar to Claims 1-7 respectively and are rejected under the same rationale as stated above for those claims.
Claims 15-20 are similar to Claims 1-6 respectively and are rejected under the same rationale as stated above for those claims.
Examiner's Note:
The Examiner cites particular pages, sections, columns, line numbers, and/or paragraphs in the references as applied to the claims above for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant fully consider the references in its entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or
disclosed by the examiner and the additional related prior arts made of record that are considered pertinent to applicant's disclosure to further show the general state of the art. The Examiner's interpretations in parenthesis are provided with the cited references to assist the applicants to better understand how the examiner interprets the prior art to read on the claims. Such comments are entirely consistent with the intent and spirit of compact prosecution.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO-892 for the relevant prior art where for example Desgarennes, US 2024/0289686 A1, teaches a generative machine learning model for providing an output that is responsive to an input and one or more environment guidelines.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVE MISIR whose telephone number is (571)272-5243. The examiner can normally be reached M-R 8-5 pm, F some hours.
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, Abdullah Al Kawsar can be reached at 5712703169. 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.
/DAVE MISIR/Primary Examiner, Art Unit 2127
Prosecution Insights