Dual-mode Internet Search System

Abstract

A computing system is disclosed that includes a processor and a memory. The memory stores instructions that, when executed by the processor, cause the processor to perform several actions. The actions include generating a prompt to be input to a generative language model. The prompt includes a conversational input set by a user. The actions further include providing the prompt as input to the generative language model and receiving a conversational output from the generative language model, where the generative language model generates the conversational output based on the prompt. Additionally, the actions include receiving an indication that a user has performed an interface mode change action and updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model. The actions further include presenting the updated SERP to a user on a client computing device.

Description

[Background technology]

[0001] background A conventional computer-implemented search engine is configured to receive a search query and infer the information-search intent of the user who issued the query (e.g., to determine whether the user wants to navigate to a particular page, whether the user intends to purchase an item or service, whether the user is searching for facts, whether the user is searching for images or videos, etc.). The search engine identifies results based on the implied information-search intent and returns a search engine results page (SERP) to the computing device employed by the user. The SERP may include links to web pages, snippets of text extracted from web pages, images, videos, knowledge cards (graphical items containing information about entities such as people, places, companies, etc.), instant answers (graphical items representing answers to questions posed in the query), widgets (e.g., a graphical calculator that can be interacted with by the user), supplemental content (e.g., advertisements related to the query), and so on.

[0002] Although search engines are frequently updated with features designed to improve the user experience (and provide increasingly relevant results to users), search engines are not well-equipped to provide some types of information. For example, search engines are not configured to provide output that requires inference about the content of a web page or output that is based on several different sources of information. For example, upon receiving the query "How many home runs did Babe Ruth hit before he turned 30?" from a user, a traditional search engine returns, among other information, a knowledge card about Babe Ruth (which may show a picture of Babe Ruth, his date of birth, etc.), alternative query suggestions ("How many hits did Babe Ruth have in his career?"), and links to web pages containing statistics. To get an answer to the question, the user must visit the web page containing the statistics and calculate the answer themselves.

[0003] In another example, upon receiving the query "Give me a list of famous people born in Seattle and Chicago," a conventional search engine returns knowledge cards about the cities Chicago and Seattle, a link to a first web page containing a list of people from Chicago, and a link to a second web page containing a list of people from Seattle. However, the search engine cannot reason about the content of the two web pages to generate a list that includes the identities of people from both Chicago and Seattle.

[0004] Relatively recently, generative language models (GLMs) (also called large language models, or LLMs) have been developed. One example of a GLM is the Generative Pre-trained Transformer 3 (GPT-3). Another example of a GLM is the BigScience Language Open-science Open-access Multilingual (BLOOM) model (also a Transformer-based model). Briefly, a GLM is configured to generate output (text in a human language, source code, music, video, etc.) based on prompts set by a user and in near real-time (e.g., within a few seconds of receiving the prompt). A GLM generates content based on the training data it has been trained on. Thus, upon receiving the prompt "How many home runs did Babe Ruth hit before he turned 30?", a GLM might output "Babe Ruth hit 94 home runs before he turned 30." In another example, in response to receiving the prompt "Give me a list of famous people born in Seattle and Chicago," a GLM may output two separate lists of people (one for Seattle and one for Chicago), where the list of people born in Chicago includes Barrack Obama. However, in both of these examples, the GLM outputs incorrect information—e.g., Babe Ruth hit over 94 home runs before turning 30, and Barrack Obama was born in Hawaii (not Chicago). Thus, both traditional search engines and GLMs are deficient in identifying and / or generating appropriate information in response to some types of user input. Summary of the Invention

[0005] overview The following is a summary of subject matter described in more detail herein. This summary is not intended to be limiting with respect to the scope of the claims.

[0006] Various techniques related to providing dual-mode search functionality by integrating GLM and search engine capabilities are described herein. Information provided as input to the GLM that is used by the GLM to generate output is called a prompt. According to some techniques described herein, the prompts used by the GLM to generate output may include: 1) user input, such as a query; and 2) information from a web page viewed by the user or information retrieved by the search engine. Prompts may also include previous dialog turns (to be described in more detail herein).

[0007] In one example, a browser on a client computing device loads a search engine web page, and the browser receives a query submitted by a user of the client computing device. The browser sends the query to a computing system running the search engine, and the search engine identifies search results and generates a search engine results page (SERP) based on the query. The search results may include web pages related to the query, knowledge cards, instant answers, entity descriptions, supplemental content, etc. The search engine returns the SERP to the browser, whereupon the SERP is displayed on the display of the client computing device when the client computing device is in SERP mode.

[0008] The user is provided with functionality that allows the user to switch from SERP mode to GLM chat mode by using one or more provided options. For example, when using a touchscreen client computing device such as a smartphone, tablet, or touchscreen computer, the user may swipe or scroll between the SERP mode interface and the GLM chat mode interface. In another embodiment, the user may tap the SERP mode graphic icon or the GLM chat mode interface graphic icon to switch between search modes. On devices without a touchscreen, the user may manipulate one or more scroll bars (e.g., on a touchpad) to scroll up or down between search modes, or may use an input device such as a mouse or directional keypad to select the graphic icon corresponding to the desired search mode.

[0009] In one example, a search engine receives the query "How many home runs did Babe Ruth hit before he turned 30?" and search results identified by the search engine include Babe Ruth's birthday and Babe Ruth's season-by-season statistics. The GLM obtains such information as part of a prompt along with the query. Because the prompt includes Babe Ruth's season-by-season home run totals, the GLM infers such data and provides output based on the information identified by the search engine as relevant to the query. Thus, the GLM may output "Babe Ruth hit 284 home runs before he turned 30." When a user switches from GLM chat mode to SERP mode (e.g., by swiping a graphical icon, scrolling, selecting, entering a voice command, etc.), the SERP interface is already populated with, for example, instant answers, entity descriptions, search results, supplemental content, etc. related to Babe Ruth. Similarly, when a user enters a query into the SERP mode interface, the search engine provides search results in the search interface, which may include links to web pages, instant answers, entity descriptions, supplemental content, etc. As the user scrolls, swipes, etc. in the GLM chat mode interface, information provided by the GLM is displayed as natural language dialogue responses.

[0010] In another example, if a user remains in GLM chat mode and performs a mode switch action (e.g., swiping or scrolling across a SERP mode screen or interface, selecting a mode graphic icon, entering a voice command indicating a desire to switch to SERP mode, etc.), the user is presented with an updated SERP that offers links to web pages, instant answers, entity descriptions, supplemental content, etc., provided by the search engine but based on the most recent query / GLM response in the GLM chat mode interface interaction. That is, in response to each new user input, query, or prompt to the GLM while in GLM chat mode, the GLM generates and sends a new query to the search engine, and the search engine updates the SERP based on the GLM query. In this way, when the user switches back to the SERP interface from the GLM chat mode interface, the SERP is up to date with the most recent instance of the GLM chat interaction.

[0011] The techniques described herein offer various advantages over conventional search engine and / or GLM techniques. Specifically, through integration with a GLM, a search engine can provide information to end users that conventional search engines cannot provide. In addition, the GLM described herein comprises information obtained by the search engine for use in generating output, thus reducing the likelihood that the GLM will issue factually incorrect or inappropriate output.

[0012] Furthermore, the described dual-mode search system integrates results from large generative text models, such as GPT3, with a traditional search engine. Results are displayed either as part of a traditional SERP or via a conversational search results page or "chat" page. Users can seamlessly transition between traditional search engine results and conversational search results by scrolling or swiping in a predetermined or user-selected direction. In another example, the transition between the search engine and conversational mode is facilitated through links within the header and body of the viewed page. Elements from traditional search results pages, such as advertisements and instant answers, can also be carried into the conversational search results page. In this way, users are allowed to seamlessly switch between traditional social search results pages and conversational search results pages. That is, users can start in traditional search and then transition to conversational mode (or vice versa) while maintaining context.

[0013] Furthermore, in another example, when changing from SERP mode to chat mode, the query set by the user and the top answer returned by the search engine can be carried over to chat mode, and the GLM response can be provided under such information, providing a seamless flow in chat mode.

[0014] The above summary presents a simplified summary to provide a basic understanding of some aspects of the systems and / or methods discussed herein. This summary is not an extensive overview of the systems and / or methods discussed herein. It is not intended to identify key / critical elements or to delineate the scope of such systems and / or methods. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later. [Brief explanation of the drawings]

[0015] BRIEF DESCRIPTION OF THE DRAWINGS [Figure 1]1 illustrates a functional block diagram of a computing system in accordance with various aspects described herein. [Figure 2] 1 illustrates a computing system with additional elements for providing seamless transitions between search modes. [Figure 3] 1 illustrates a schematic diagram depicting a GUI of an operating system installed on a computing device illustrating a SERP mode interface in accordance with various aspects described herein. [Figure 4] 1 shows a schematic representation of the GUI of an operating system installed on a computing device showing a GLM chat mode interface. [Figure 5] 1 is an illustration of a GUI on a communication device in SERP mode according to one or more aspects described herein. [Figure 6] 1 illustrates a GUI on a communication device in GLM conversation mode according to various embodiments described herein. [Figure 7] 1 illustrates a flow diagram for providing dual-mode search functionality in a computing system according to one or more aspects described herein. [Figure 8] 1 illustrates a flow diagram for providing dual-mode search functionality on a client device according to one or more aspects described herein. [Figure 9] 1 is a high-level illustration of an exemplary computing device that may be used in accordance with the systems and methodologies disclosed herein. DETAILED DESCRIPTION OF THE INVENTION

[0016] Detailed Description Various techniques relating to providing dual-mode search functionality on a computing device will now be described with reference to the accompanying drawings, wherein like reference numerals are used to refer to like elements throughout the specification. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more aspects. However, it will be apparent that such aspects may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form to facilitate describing one or more aspects. Furthermore, it will be understood that functionality described as being performed by several system components may be performed by multiple components. Similarly, for example, one component may be configured to perform functionality described as being performed by multiple components.

[0017] Furthermore, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless otherwise specified or clear from the context, the phrase "X employs A or B" is intended to mean any of the natural inclusive permutations. That is, the phrase "X employs A or B" is satisfied by any of the following cases: X employs A; X employs B; or X employs both A and B. Additionally, the articles "a" and "an," as used in this application and in the appended claims, should be generally construed to mean "one or more" unless otherwise specified or clear from the context that the singular form is directed.

[0018] Furthermore, as used herein, the terms "component," "system," "engine," and "module" are intended to encompass computer-readable data storage configured with computer-executable instructions that, when executed by a processor, cause some functionality to be performed. The computer-executable instructions may include routines, functions, or the like. It should also be understood that a component or system may be localized on a single device or distributed across several devices. Furthermore, as used herein, the term "exemplary" is intended to mean serving as an illustration or example of something, but is not intended to indicate preference.

[0019] Described herein are various techniques related to providing dual-mode search functionality through the use of a search engine and a generative language model (GLM), also known as a large-scale language model (LLM). The described systems and methods allow for a quick and seamless transition between the "SERP" method of searching using the search engine and a GLM chat mode, where the user is provided with an online chat interaction experience. Furthermore, when switching from the GLM chat mode to the SERP mode, the SERP is automatically updated to display information related to the most recent prompt / response in the GLM interaction.

[0020] 1, there is shown a functional block diagram of a computing system 100 in accordance with various aspects described herein. Although shown as a single system, it will be understood that computing system 100 may include several different server computing devices, may be distributed across several data centers, etc. Computing system 100 is configured to retrieve information based on a query established by a user, and is further configured to provide the retrieved information to a GLM as part of a prompt.

[0021] A client computing device 102 operated by a user (not shown) is in communication with computing system 100 via network 104. Client computing device 102 may be any suitable type of client computing device, such as a desktop computer, a laptop computer, a tablet (slate) computing device, a video game system, a virtual reality or augmented reality computing system, a mobile phone, a smart speaker, or other suitable computing device.

[0022] Computing system 100 includes processor 106 and memory 108, where memory 108 includes instructions executed by processor 106. Specifically, memory 108 includes search engine 110 and GLM 112, where the operation of search engine 110 and GLM 112 is described in more detail below. Computing system 106 also includes data storages 114-122, where data storages 114-122 store data accessed by search engine 110 and / or GLM 112. Specifically, data storages 114-122 include web index data storage 114, instant answer data storage 116, knowledge graph data storage 118, supplemental content data storage 120, and interaction history data storage 122. Web index data storage 114 includes web indexes that index web pages by keywords contained in or associated with the web pages. The instant answer data storage 116 includes indices of instant answers that are indexed by queries, query terms, and / or terms that are semantically similar or equivalent to the queries and / or query terms. For example, the instant answer "2.16 meters" may be indexed by the query "Shaquille O'Neal's height" (and semantically similar or equivalent queries such as "How tall is Shaquille O'Neal").

[0023] The knowledge graph data storage 118 contains a knowledge graph, where the knowledge graph includes data structures about entities (people, places, things, etc.) and their relationships to each other, thereby representing the relationships between the entities. The search engine 110 may use the knowledge graph in connection with presenting entity cards on a search engine results page (SERP). The supplemental content data storage 120 contains supplemental content that may be returned by the search engine 110 based on a query.

[0024] The interaction history data storage 122 includes interaction history, where the interaction history includes interaction information regarding the user and the GLM 112. For example, the interaction history may include a user's: identifiers of the conversation undertaken between the user and the GLM 112; input provided by the user to the GLM 112 for multiple dialogue turns during the conversation; responses in the conversation generated by the GLM 112 in response to input from the user; queries generated by the GLM during the conversation that are used by the GLM 112 to generate responses; etc. Additionally, the interaction history may include context captured by the search engine 110 during the conversation; for example, with respect to a conversation, the interaction history 122 may include content from SERPs generated based on queries set by the user and / or the GLM 112 during the conversation, content from web pages identified by the search engine 110 based on queries set by the user and / or the GLM 112 during the conversation, etc. Data storages 114-122 are presented to show a representative sample of the types of data accessible to search engine 110 and / or GLM 112; it will be understood that there are many other data sources accessible to search engine 110 and / or GLM 112 (e.g., data storage containing real-time financial information, data storage containing real-time weather information, data storage containing real-time sports information, data storage containing images, data storage containing video, data storage containing maps, etc.) that may be available to search engine 110 and / or GLM 112.

[0025] The search engine 110 includes a web search module 124, an instant answer search module 126, a knowledge module 128, a supplemental content search module 130, and a SERP builder module 132. The web search module 124 is configured to search the web index data storage 114 based on queries received by a user, queries generated by the search engine 110 based on queries received by a user, and / or queries generated by the GLM 112 based on the user's interactions with the GLM 112. Similarly, the instant answer search module 126 is configured to search the instant answer data storage 116 based on queries received by a user, queries generated by the search engine 110 based on queries received by a user, and / or queries generated by the GLM 112 based on the user's interactions with the GLM 112. The knowledge module 128 is configured to search the knowledge graph data storage 118 based on queries received by a user, queries generated by the search engine 110 based on queries received by a user, and / or queries generated by the GLM 112 based on the user's interactions with the GLM 112. Similarly, the supplemental content search module 130 is configured to search the supplemental content data storage 120 based on queries received by a user, queries generated by the search engine 110 based on queries received by a user, and / or queries generated by the GLM 112 based on the user's interactions with the GLM 112.

[0026] The SERP builder module 132 is configured to build a SERP based on information identified by searches performed by modules 124-130; for example, the SERP may include links to web pages identified by the web search module 124, instant answers identified by the instant answer search module 126, entity cards (containing information about entities) identified by the knowledge module 128, and supplemental content identified by the supplemental content search module 130. Additionally, the SERP may include widgets, cards depicting current weather, etc. The SERP builder module 132 may also generate structured, semi-structured, and / or unstructured data representing the content of the SERP or portions of the content of the SERP. For example, the SERP builder module 132 generates a JSON document containing information retrieved by the search engine 110 based on one or more searches performed across the data storages 114-120 (or other data storages). In one example, the SERP builder module 132 generates data in a structure / format used as part of a prompt by the GLM 112.

[0027] As discussed above, the operation of the search engine 110 is improved based on the GLM 112, and the operation of the GLM 112 is improved based on the search engine 110. For example, the search engine 110 can provide output that the search engine 110 was not previously able to provide (e.g., based on output generated by the GLM 112), and the GLM 112 is improved by using information obtained by the search engine 110 to generate the output (e.g., information identified by the search engine 110 can be included as part of the prompts used by the GLM 112 to generate the output). Specifically, because the search engine 110 is associated with several years of design to curate information sources and ensure their accuracy, the GLM 112 generates results based on information obtained by the search engine 110 that have a higher likelihood of being accurate compared to results generated by the GLM 112 that are not based on such information.

[0028] Continuing with reference to Figure 1, Figure 2 illustrates a computing system 100 that includes a processor 106 and a memory 108 as described with respect to Figure 1, where the memory 108 includes instructions executed by the processor 106. Specifically, the memory 108 includes a search engine 110 and a GLM 112, where the operation of the search engine 110 and the GLM 112 is described in more detail above with respect to Figure 1. The computing system 106 also includes data storage 114-122 that stores data accessed by the search engine 110 and / or the GLM 112 as described above.

[0029] The search engine 110 includes a web search module 124 , an instant answer search module 126 , a knowledge module 128 , a supplemental content search module 130 , and a SERP builder module 132 .

[0030] 1 , the SERP builder module 132 includes a SERP / GLM transition module 202 and a SERP update module 204. The SERP / GLM transition module 202 receives indications of SERP / GLM transition actions taken on the client device 102. The SERP / GLM transition actions indicate a user's desire to switch between a SERP mode of search and a GLM chat mode (or vice versa). For example, if a user is in GLM chat mode and then swipes toward SERP mode, selects the SERP mode graphic icon, enters a voice command to switch to SERP mode, etc., the client device sends an indication of the user's desire to switch modes to the computing system 100. The SERP / GLM transition module 202 detects the transition action, and the SERP update module updates the SERP provided to the client device to include information related to the most recent query / response provided in GLM chat mode (e.g., traditional query mode, which returns search results, instant answers, entity descriptions, etc.) to include instant answers, entity descriptions, search results, supplemental content, etc. Conversely, if a user wishes to switch from GLM chat mode to SERP mode, the indication is detected by the SERP / GLM transition module 202, and the generative language model 112 takes precedence over the search system 110 for providing interactive responses to the client device and GLM chat mode.

[0031] It will be understood that the various databases described herein with respect to Figures 1 and 2 store cached information (e.g., cached web pages, or other data sources) for responding to chat queries and / or providing search results in the SERPs or other information provided to client devices. Cached web pages are periodically updated and / or invalidated to maintain current source data.

[0032] 3, a schematic diagram depicting a GUI generated by computing system 100 is shown, where GUI 312 includes a query field 314 into which a user may type a query to conduct a search. The query is sent to a search engine (not shown), which returns instant answers 316 (if applicable), entity descriptions 318 (if applicable), links to web pages 320 identified by the search engine as related to the query, and supplemental content 322. Instant answers 316 are, for example, answers provided by the search engine in response to a query without the user having to navigate away from SERP 312. Instant answers may be answers that have been previously validated and cached in response to the same query.

[0033] It will be understood that the depicted orientation of the query field 314, instant answer 316, entity description 318, search results 320, and supplemental content 322 relative to one another is presented for illustrative purposes only and is not intended to limit the specific placement of these elements within the SERP 312. For example, the query field 314 may be presented below the results 320 and entity description 318. In another example, the position of the instant answer 316 may be swapped with the position of the entity description 318. In another example, the query field 314, instant answer 316, entity description 318, results 320, and supplemental content 322 may be presented as a vertical stack of fields in any order.

[0034] A slider bar 330 is provided for scrolling up or down on the main display area 304. For example, a user may use the slider bar 330 to navigate up or down within the results field through various results (labeled A, B, C, D...) and / or one or more images 332. Similarly, the user may use slider bar 328 to navigate up or down through supplemental content results (labeled as C', C", ...) and / or one or more supplemental images 334. In one embodiment, the user may hover pointer 310 over a particular result in results field 320, and the system will search for and present additional content in supplemental content field 322 related to the result over which the pointer is hovered. In another embodiment, when a user hovers their pointer over a particular result, a pop-up window is displayed showing the source of information and / or supplemental information, such as advertisements (images or videos) presented on the source page. In the example of FIG. 3, the user hovered their pointer over a link to web page C, and the system searched for supplemental content C' and C" related to the content being hovered over.

[0035] When a user decides to switch from SERP mode to chat mode (also referred to herein as "conversation mode"), the user may use a slider bar 330 to navigate upward to a generative language model (GLM) conversation mode interface 336. Additionally or alternatively, a graphic icon 338 may be provided that, when selected by the user, causes the screen to scroll upward or otherwise switch to the GLM conversation mode interface 336. It will be understood that the swipe direction required to switch between SERP and GLM conversation mode is not limited to an upward direction, but rather can be a downward swipe, a rightward swipe, or a leftward swipe, as will be understood by those skilled in the art.

[0036] In another embodiment in which computing device 102 includes a touchscreen, a user may simply swipe up or down using a finger, stylus, or other device. In this embodiment, slider bar 330 is optional. Alternatively, slider bar 330 may be held or displayed when the user's finger or stylus touches the screen within one of field 320, entity description field 318, supplemental content field 322, or main display area 304. When the user releases the touchscreen, the respective slider bar disappears.

[0037] In another embodiment, the client computing device 102 includes a microphone (not shown) through which a user can initiate voice commands to search and / or to switch between search modes (SERP and GLM). In one example, when a user initiates a voice command, the search mode defaults to conversation mode or chat mode. In another example, the search mode defaults to SERP mode. In yet another example, the user is allowed to configure the default search mode according to user preferences.

[0038] Referring now to FIG. 4 , a schematic diagram depicting another view of GUI 300 is shown. GUI 300 depicts a conversation mode interface 336 to which a user navigates from SERP interface 312 via slider bar 330, via chat mode graphic icon 338, via touchscreen functionality, via an input device such as a mouse or directional pad on a keyboard, etc. GLM conversation mode interface 336 includes an input field 402 that can be selected via pointer 310, a finger or stylus in the case of a touchscreen, or any other suitable means. The user enters conversational input into input field 402 for transmission to GLM 112, and once transmitted, the conversational input appears in conversation field 404. Conversation field 404 optionally includes a slider bar 406 that allows the user to navigate up and down through the chat dialogue. User queries are sent to GLM 112 ( FIG. 1 ), which returns natural language responses in response to the conversational input. The user is then allowed to respond to the natural language response provided by the search engine as if conversing with another human. The GLM 112 then provides a second natural language response to the user, and the conversation continues. Meanwhile, the user may hover the pointer 310 over any of the natural language responses provided by the GLM 112, and the GLM 112 may form and submit a query to return supplemental content 322 for presentation within or adjacent to the GLM conversation mode interface 336 on the GUI 300. The supplemental content 322 typically includes one or more selectable links to articles or web pages related to the natural language response over which the pointer 310 is hovered, and may include one or more supplemental images 334. A slider bar 328 is provided within the supplemental content field 322 and allows the user to scroll through the supplemental content and / or images 334 within the supplemental content field.

[0039] In one embodiment, the GUI also includes selectable SERP graphic icons that, when selected or otherwise activated by the user, cause the system to return to the SERP interface 312. Additionally or alternatively, the user may employ a scroll bar 330 to scroll back into the SERP interface 312. In another embodiment in which the GUI 300 is displayed on a touchscreen device, the user may simply use a finger or stylus to scroll back into the SERP interface 312.

[0040] According to another feature, when the user returns to the SERP interface 312, the instant answers 316, entity descriptions 318, query results 320, and supplemental content 322 (see FIG. 3 ) are populated with information related to the last user query entered during the conversation in the GLM conversation mode interface 336. That is, the SERP is updated to reflect the results of the most recent query made in the GLM conversation mode interface 336. In this way, the user is allowed to switch back and forth between the SERP interface 312 and the GLM conversation mode interface 336 while being seamlessly and continuously provided with updated search result information regardless of which interface the user is currently using.

[0041] In another embodiment, the client computing device 102 includes a microphone (not shown) through which a user can initiate voice commands to search and / or to switch between search modes (SERP and GLM). In one example, when a user initiates a voice command, the search mode defaults to conversation mode or chat mode. In another example, the search mode defaults to SERP mode. In yet another example, the user is allowed to configure the default search mode according to user preferences.

[0042] Referring now to FIG. 5 , a GUI 500 is illustrated on a communication device 502, such as a tablet, cell phone, or smartphone, in accordance with one or more aspects described herein. The GUI 500 includes a SERP interface 504 that includes a query field 506, an instant answer field 508 (if applicable), an entity description field 510 (if applicable), and a results field 512. Displayed within the results field 512 are one or more links and optionally one or more images 514 to web pages (labeled A-D) returned in response to the query entered into the query field 506. The user clicks on one of the returned results A-D, and the device displays received information related to the selected result. In addition, the system retrieves supplemental content 516 (e.g., additional articles, hyperlinks, images, advertisements, etc.) related to the selected result and displays the supplemental content on the SERP interface 504.

[0043] It will be understood that the particular order of the query field 506, instant answer 508, entity description 510, results field 512, and supplemental content field 514 is not limited to that depicted in Figure 5, but rather these elements may be arranged in any order. Furthermore, the depicted elements in the SERP interface 504 are not limited to a stacked arrangement as shown in Figure 5, but rather may be arranged side by side, such as in a grid arrangement.

[0044] The communication device 502 further includes a microphone 518 and one or more speakers 520, through which a user can input voice commands and receive audio from the communication device. For example, a user can initiate a query by activating the microphone and saying the word "query" or "question" (followed by a word or phrase that the user might otherwise manually enter into the query field 506). The results field 512 can be populated with results (e.g., hyperlinks, article titles, images 512, etc.) in response to the user's voice query. In another embodiment, the results can be read and presented to the user as audio output via the speaker 520.

[0045] In another embodiment, a voice-activated graphical icon (not shown) may be provided in the query field 506 or elsewhere in the SERP interface 504. In response to a user's selection (e.g., tapping or long pressing) of the voice-activated graphical icon, the user is prompted to begin speaking so that they can speak a natural language query into a microphone 518. One or more of the returned instant answers 508, entity descriptions 510, results 512, and / or supplemental content 514 may be presented to the user as audio output via one or more speakers 520.

[0046] When a user wishes to switch from the SERP interface 504 to the GLM conversation mode interface 522, the user scrolls up on the GUI 500 (e.g., by using a finger or stylus to activate a touchscreen). Additionally or alternatively, a chat mode graphical icon (not shown) may be presented on the SERP interface 504 or elsewhere on the GUI 500, where the graphical icon may be selected or activated by the user to switch to the GLM conversation mode. In yet another embodiment, the user may give a voice command, such as "chat mode" or some other suitable voice command, to switch from the SERP mode to chat mode.

[0047] 6, shown on the GUI 500 of the communication device 502 is a GLM conversation mode interface 522 according to various embodiments described herein. The GLM conversation mode interface 522 includes an input field 602 into which a user may type or speak a conversational input. The GLM conversation mode interface 522 also includes a conversation field 604 that shows the user's initial conversational input (Input 1) and the system's natural language response to that input (Response 1) in response to the conversational input being sent to the GLM 112. The user query and the natural language response generated by the system are displayed to the user in the conversation field 604 as a dialogue. An example of a conversation / dialogue that may be displayed in the conversation field 604 is provided below.

[0048] Query 1: What state is Ann Arbor in? Response 1: Ann Arbor is located in the state of Michigan in the United States. Query 2: Tell me more. Response 2: Ann Arbor is a city in the southeastern region of Michigan, approximately 35 miles (56 km) west of Detroit. Ann Arbor is the county seat of Washtenaw County and is known as the home of the University of Michigan, one of the oldest and most prestigious public universities in the United States. Query 3: What SAT score does the University of Michigan require? Response 3: The University of Michigan requires students to submit SAT scores as part of their application. For the SAT, the mid-50s range for 2025 was 1340-1470.

[0049] As can be seen, the responses generated by the system take into account the context of the conversation. For example, if the user references "university" in query 3, the system infers that the user is referring to the University of Michigan based on the context of response 2. Communication device 502 also includes a microphone 518 and one or more speakers 520 that allow the user to speak queries and listen to responses in conversation, as described above with respect to FIG. 5.

[0050] The system can also generate supplemental content 606 for display within the GLM conversation mode interface 522 or elsewhere on the GUI 500. The supplemental content 606 is identified / searched for using the conversation context and may include additional links, images, selectable graphic icons, etc. that the user can click on for additional information. For example, the content may include, but is not limited to, links to one or more hotels in the Ann Arbor area, links to restaurants in Ann Arbor, links to buy tickets to University of Michigan sporting events, etc.

[0051] If the user wishes to return to the SERP mode interface 504, the user simply scrolls down on GUI 500. In another embodiment, the user is allowed to use voice commands to switch between SERP mode and GLM conversation mode. When the user returns to the SERP interface, the instant answers and results fields are populated with information related to the SAT score requirements at the University of Michigan, while the entity description field presents information about the University of Michigan itself. The supplemental content field is populated with supplemental content similar to that presented on the GLM conversational mode interface.

[0052] There are various other features contemplated with reference to a system integrating a search engine and a GLM. For example, as previously indicated, the GLM 112 may generate conversational output based on conversational input. In one example, the GLM 112 may analyze the conversational input and / or the conversational output to generate a variety of outputs. For example, the GLM 112 may generate query suggestions suitable for sending to the search engine 110 so that the GLM 112 may prompt the user to switch to search engine mode. For example, based on the conversational input "What SAT score do colleges require?", the GLM 112 may generate several queries (such as "Places near me where the SAT test is held," "Dates for the SAT test," among others) configured to be received by the search engine 110. The GLM 112 may assign hyperlinks to text in the conversational input and / or text in the conversational output, where hovering over the hyperlink may present one or more query suggestions.

[0053] In another example, as noted above, the search engine 110 is configured to output instant answers and / or knowledge cards, if applicable. For example, in response to receiving the query "Company A stock price," the search engine 110 generates an instant answer identifying Company A's stock price. The GLM 112 may be provided with the content of the query and / or instant answer submitted by the user and may generate further query suggestions and / or conversational input suggestions based on the content of the query and / or instant answer submitted by the user. The query suggestions and / or conversational input suggestions may be presented along with the instant answer to visually indicate that the query suggestions and / or conversational input suggestions correspond to the instant answer. An example query suggestion is "Who is the CEO of Company A?" and an example conversational input suggestion is "Explain the business differences between Company A and Company B." Additionally, graphical indications may be presented to identify which suggestions are conversational suggestions and which are query suggestions used by the search engine 110 to identify search results. When a conversational input suggestion is selected, such suggestion is used as input by the GLM 112 to generate a conversational output, and the context may switch to a conversational mode (e.g., GUI features related to the conversational mode are presented). While the above example related to instant answers, it will be understood that similar features may be employed in connection with knowledge cards, widgets, and / or supplemental content.

[0054] In yet another example, rather than having two separate and distinct interfaces for the conversational mode and the search engine mode, the interfaces may be integrated with one another. For example, a conversational interface may be presented within a sidebar. In yet another example, when input is received, a classifier may identify whether the input set by the user is conversational in nature and therefore to be provided to the GLM 112 or whether the input is more appropriate for placement in a search engine. For example, the input "Company A's stock price" is typically more appropriate as a query to be issued to a search engine, while the input "Explain the differences between Company A's and Company B's businesses" is typically more appropriate as conversational input to be issued to the GLM 112. In the former case, the query is provided to the search engine 110, and the search engine 110 performs the search; for example, the search engine 110 provides an instant answer identifying the current stock price of Company A. In the latter case, the conversational input is provided to the GLM 112, and the GLM 112 generates a conversational output based on such input. In one example, the interactive output may be provided within a GUI that integrates conversational and search modes, resulting in an interactive output that most closely resembles a knowledge card about companies A and B.

[0055] In yet another example, when generating the conversational output, the GLM 112 may identify sources of information contained within the output and may generate hyperlinks corresponding to those sources and / or queries that can be used by the search engine 110 to search the sources. For example, and referring to the conversational input "Explain the differences between Company A's business and Company B's business," the GLM 112 may generate the following output: "Company A's business is to make Type 1 widgets that are primarily manufactured in location C. In contrast, Company B's business is to make Type 2 widgets that are primarily manufactured in location D. Company A had more revenue than Company B in 2022, but Company B's profit margins are higher than Company A's."

[0056] GLM 112 may identify sources used by GLM 112 to identify the types of widgets made by certain companies, where the widgets are manufactured, revenue figures for the companies, and profit margins associated with the companies. Upon hovering over text (e.g., "Revenue"), GLM 112 may present query suggestions to search for information about company revenues and / or identifiers of web pages containing revenue information for Company A and / or Company B. Thus, users can be assured that the information provided by GLM 112 is accurate and up-to-date.

[0057] While the examples presented above are referred to as generic search engines, it is contemplated that the techniques described herein are applicable to enterprise search engines. For example, an enterprise search engine may be configured to search on documents specific to an enterprise, which may include internal web pages and word processing documents, slide shows, etc. Similar to that referenced above, enterprise searches may be integrated with a GLM such that interactive output related to enterprise documents can be generated based on interactive input and displayed in a GUI related to chat and / or in a GUI with chat integrated with traditional search.

[0058] 7-8 illustrate a methodology for providing dual-mode search functionality that allows a user to seamlessly switch between a generative language model conversation or chat mode and a search engine results page mode according to one or more embodiments described herein. While the methodology is shown and described as a series of acts performed in sequence, it will be understood and appreciated that the methodology is not limited by the order of the sequence. For example, some acts may occur in a different order than described herein. In addition, acts may occur simultaneously with other acts. Furthermore, in some instances, not all acts may be required to implement the methodology described herein.

[0059] Furthermore, the acts described herein may be computer-executable instructions that may be performed by one or more processors and / or stored on a computer-readable medium or media. Computer-executable instructions may include routines, subroutines, programs, threads of execution, etc. Furthermore, the results of the acts of the methodologies may be stored in a computer-readable medium, displayed on a display device, etc.

[0060] Referring now to FIG. 7 , a flow diagram is depicted depicting a method 700 of providing dual-mode search functionality within a computing system according to one or more aspects described herein. At 702, a user query is received in GLM chat mode. At 704, a GLM response dialogue is generated and returned (e.g., via a generative language model (see FIG. 1 )). At 706, the dialogue context is analyzed. At 708, supplemental content is identified based on the dialogue context and returned to the user device for display to the user. While the GLM chat dialogue is progressing at 702-708, the SERP is simultaneously populated at 710 with information responsive to the final query and / or response in the GLM chat. At 712, a determination is made as to whether an interface mode change indication has been detected. The interface mode change indication may be triggered, for example, by a user swiping, scrolling, selecting a mode change graphic icon, entering a voice command to change modes, etc. If the test at 712 indicates that no interface mode change indication has been detected, the method returns to 702 for continued GLM chat mode operation and receipt of subsequent queries.

[0061] If it is determined at 712 that an interface mode change indication has been detected, the method proceeds to 714, where query and search result operation continues in SERP mode. The user's device may now present a SERP populated with information responsive to the final GLM Chat query and / or response. At 716, a determination is made whether an interface mode change indication has been detected. If no, the method returns to 714 for continued operation in SERP mode. If the determination at 716 indicates that a mode change indication has been detected, the method proceeds to 718, where system operation returns to GLM Chat mode.

[0062] Referring now to FIG. 8 , a flow diagram depicts a method 800 for providing dual-mode search functionality on a client device according to one or more aspects described herein. At 802, a user query in a GLM chat mode is detected by the client device and transmitted (e.g., over a network) to a computing system. At 804, a GLM response dialogue is received, e.g., from a generative language model on the computing system, and displayed on the client device. At 806, supplemental content based on the dialogue context is received and displayed on the client device. At 808, concurrently with the GLM chat dialogue occurring at 802-806, the SERP is updated with information related to the final query and / or response in the GLM chat dialogue. At 810, a determination is made as to whether an interface mode change action is detected on the client device. The interface mode change action may be, for example, a user swiping, scrolling, selecting a mode change graphic icon, entering a voice command to change modes, etc. If the determination at 810 indicates that no interface mode change action has been detected, the method returns to 802 for continued GLM chat mode operation and receipt of a subsequent query. If a determination at 810 indicates that an interface mode change indication has been detected, the method proceeds to 812, and an updated SERP is presented to the user on the client device. At 814, query and search result operation continues in SERP mode. At 816, a determination is made whether an interface mode change action has been detected. If no, the method returns to 814 for continued operation in SERP mode. If the determination at 816 indicates that an interface mode change action has been detected, the method proceeds to 818, where client device operation and display return to GLM chat mode.

[0063] With continued reference to Figures 1-8, various additional contemplated features and aspects are described below. In one embodiment, conversation mode may be entered from any search endpoint (including, but not limited to, a web search, a multimedia search, a shopping portal, videos, maps, news, artwork, etc.). For example, a user may click a "tip" or graphic icon on the display screen to enter GLM conversation mode while in SERP mode, but may be presented with another graphic icon in conversation mode that can be selected to re-enter SERP mode. When using a touchscreen computer or mobile device, a user may swipe up / down or left / right to switch between SERP mode and conversation mode, depending on the designated orientation of the mode interface. In a related embodiment, the swipe direction for changing between SERP conversation mode and GLM conversation mode is user-configurable.

[0064] In another embodiment, selectable graphic icons may be presented in a sidebar on a displayed page, in a sidebar or other panel in a web browser, and / or elements displayed on a page may be wrapped with a conversation. The system may also be configured to ask the user questions as part of a conversation (e.g., ask open-ended clarifying questions) in addition to providing suggested tips or icons with fixed responses.

[0065] Other features include voice-driven search, visual question answering on visual content, bot labeling annotations on objects on the page (icons plus annotations), voice annotations on user interactions with page elements, adaptive generation based on user interaction and attention, dialogue-driven interaction with content on SERP pages, wrapper or right rail-overlay, interaction with other UIs for other content (email exploration, searching corporate document repositories, etc.).

[0066] Additional features may include: dialogue-driven interaction with dynamically generated web pages; dynamic layout rearrangement; dynamic compilation of elements into the dialogue history; full-page transitions from dialogue elements in a conversation; weather elements in a conversation; ability to click on an answer card to switch back to a full-page portal / details / non-mini version of the element; ability to switch between in-conversation element and full experience mode; etc. For example, a user may switch from a shopping answer to a full-page shopping page. Other features include: answer / exploded view; ability to set aside an answer during a conversation; ability to pin an answer; ability to put an answer in a new browser, etc. Additionally, an "expand" button may be provided for answers that have an alternate expansion mode to switch to.

[0067] Additionally, options for multi-page conversations are provided: for example, a given model state may include dialogue / interactions across multiple tabs / pages / explorations.

[0068] Other features include the ability to have the content of an entire news article shown as an answer added to the dialogue context. For example, the system may fetch a news article but show only the headline and image of the news article. In this scenario, the user may pose a next question that utilizes the full content of the news article and / or prioritize fetching content to generate the next response.

[0069] In another embodiment, search results in conversational mode may be delivered as web result answer cards and / or semantic summary answer cards.

[0070] The described systems and methods also provide the ability to share conversations with others, allow multi-party conversations, save conversations for later resumption, bookmark conversations, save the entire dialogue history for later review, timestamp dialogues / conversations so that the next response can leverage recent dialogue history across multiple windows / conversations, share conversation turns widely (e.g., on social media), integrate mixed-mode external content / dialogue into enterprise chat applications, provide upsells to another application experience, and the like.

[0071] In another example, the described systems and methods facilitate providing a "new tab page" that includes one or more of a "what's new" summary, asynchronous updates on what's happening with user data, user interests, what's happening in the world, etc. Email can be another experience where mini-response / extended / full-page modes are provided. For email, sub-responses can be individual emails, information about people, person responses / cards, etc. The "extended" mode can launch a new window or tab for compiling / sending an email response. The response can include a short list of related emails or SharePoint items. The extended mode can also transition to a full-page Word document of document results. Also contemplated is an option to switch back to a conversation with the mini-response mode.

[0072] Referring now to FIG. 9 , a high-level diagram of an exemplary computing device 900 that can be used in accordance with the systems and methodologies disclosed herein is shown. For example, the computing device 900 can be a client computing device having an operating system stored thereon, where the operating system provides dual-mode SERP / GLM search functionality. As another example, the computing device 900 can be a server computing system that provides dual-mode SERP / GLM search functionality. The computing device 900 includes at least one processor 902 that executes instructions stored in memory 904. The instructions can be, for example, instructions for implementing functionality described as being performed by one or more components discussed above or instructions for implementing one or more of the methods described above. The processor 902 can access the memory 904 via a system bus 906. In addition to storing executable instructions, the memory 904 can also store content, graphic icons, profile information, etc.

[0073] Computing device 900 additionally includes data storage 908 accessible by processor 902 via system bus 906. Data storage 908 may include executable instructions, graphic icons, profile information, content, etc. Computing device 900 also includes an input interface 910 that allows external devices to communicate with computing device 900. For example, input interface 910 may be used to receive instructions from an external computer device, from a user, etc. Computing device 900 also includes an output interface 912 that interfaces computing device 900 with one or more external devices. For example, computing device 900 may display text, images, etc. via output interface 912.

[0074] It is contemplated that external devices communicating with computing device 900 via input interface 910 and output interface 912 may be included within an environment providing virtually any type of user interface with which a user may interact. Examples of user interface types include graphical user interfaces, natural user interfaces, etc. For example, a graphical user interface may receive input from a user employing input devices such as a keyboard, mouse, remote control, etc., and provide output on an output device such as a display. Furthermore, a natural user interface may allow a user to interact with computing device 900 in a manner free of the constraints imposed by input devices such as a keyboard, mouse, remote control, etc. Rather, a natural user interface may rely on speech recognition, touch and stylus recognition, both on-screen and adjacent-to-screen gesture recognition, air gestures, head and eye tracking, voice and speech, vision, touch, gestures, machine intelligence, etc. Additionally, although shown as a single system, it will be understood that computing device 900 may be a distributed system. Thus, for example, several devices may be in communication via a network connection and collectively perform the tasks described as being performed by computing device 900 .

[0075] The various functions described herein may be implemented in hardware, software, or any combination thereof. If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. A computer-readable medium includes a computer-readable storage medium. A computer-readable storage medium may be any available storage medium that can be accessed by a computer. By way of example, and without limitation, such computer-readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code means in the form of instructions or data structures and that can be accessed by a computer. As used herein, disk and disc include compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs (BDs), where disks typically replicate data magnetically and discs typically replicate data optically with a laser. Furthermore, propagated signals are not included within the scope of computer-readable storage media. Computer-readable media also includes communication media, which include any medium that facilitates transfer of a computer program from one place to another. A connection may be, for example, a communications medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technology (such as infrared, radio, and microwave), the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technology (such as infrared, radio, and microwave) are included in the definition of communications media. Combinations of the above should also be included within the scope of computer-readable media.

[0076] Alternatively or additionally, the functionality described herein may be performed at least in part by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0077] Described herein are various techniques by way of at least the following examples.

[0078] (A1) In one aspect, a computing device is described herein. The computing device includes a processor and a memory storing instructions that, when executed by the processor, cause the processor to perform several actions. The actions include generating a prompt to be input to a generative language model, where the prompt includes a conversational input set by a user. The actions further include providing the prompt as input to the generative language model. The actions also include receiving a conversational output from the generative language model, where the generative language model generates the conversational output based on the prompt. Additionally, the actions include receiving an indication that a user has performed an interface mode change action. The actions further include updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model. The actions also include presenting the updated SERP to a user on a client computing device.

[0079] (A2) In some embodiments of the computing device of (A1), the acts further include determining a dialogue context including conversational input and output, and receiving and displaying supplemental content based on the dialogue context.

[0080] (A3) In some embodiments of at least one of the computing devices of (A1)-(A2), the acts further include receiving and displaying from the generative language model one or more selectable query suggestions generated by the generative language model based on the conversational input and output.

[0081] (A4) In some embodiments of the computing device of (A3), the act further includes prompting the user to switch to SERP mode in response to a user selection of a selectable query suggestion among the one or more selectable query suggestions.

[0082] (A5) In some embodiments of at least one computing device of (A1)-(A4), the act further includes receiving and displaying one or more selectable hyperlinks corresponding to one or more information sources included in the interactive output, the hyperlinks being usable by a search engine when selected to search the one or more information sources.

[0083] (A6) In some embodiments of at least one of the computing devices of (A1)-(A5), the action further includes receiving an indication that the user has performed an additional interface mode change action and resuming display of interactive input and output.

[0084] (A7) In some embodiments of at least one of the computing devices of (A1) through (A6), the interface mode change action includes at least one of swiping on a graphical user interface to switch between a SERP mode and a conversation mode, selecting a selectable graphical icon, a voice command, and manipulating a slider bar.

[0085] (B1) In another aspect, a computing system is described herein. The computing system includes a processor and a memory storing instructions that, when executed by the processor, cause the processor to perform several actions. The actions include receiving a prompt as input to a generative language model, where the prompt includes a conversational input set by a user. The actions further include generating and displaying a conversational output from the generative language model, where the conversational output is generated based on the prompt. The actions also include receiving an indication that a user has performed an interface mode change action. The actions further include updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model. Additionally, the actions include providing the updated SERP to a user on a client computing device.

[0086] (B2) In some embodiments of the computing system of (B1), the acts further include determining a dialogue context including conversational input and output, and retrieving and displaying supplemental content based on the dialogue context.

[0087] (B3) In some embodiments of at least one of the computing systems of (B1)-(B2), the acts further include generating and displaying one or more selectable query suggestions with the generative language model, the selectable query suggestions being based on the conversational input and output.

[0088] (B4)(B3) In some embodiments of the computing system, the act further includes prompting the user to switch to SERP mode in response to a user selection of one of the one or more selectable query suggestions.

[0089] (B5) In some embodiments of at least one of the computing systems of (B1) through (B4), the actions further include generating and displaying one or more selectable hyperlinks corresponding to one or more information sources included within the interactive output, the selectable hyperlinks being usable by a search engine when selected to search the one or more information sources.

[0090] (B6) In some embodiments of at least one of the computing systems of (B1) through (B5), the action further includes receiving an indication that the user has performed an additional interface mode change action and resuming display of interactive input and output.

[0091] (B7) In some embodiments of at least one of the computing systems of (B1) through (B6), the interface mode change action includes at least one of swiping on a graphical user interface, selecting a selectable graphical icon, a voice command, and manipulating a slider bar to switch between a SERP mode and a conversation mode.

[0092] (C1) In another aspect, a method performed by a computing system is described herein. The method includes receiving a prompt as input to a generative language model, where the prompt includes a conversational input set by a user. The method further includes generating and displaying a conversational output from the generative language model, where the conversational output is generated based on the prompt. The method also includes receiving an indication that the user has performed an interface mode change action. Further, the method includes updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model. The method also includes providing the updated SERP to the user on a client computing device.

[0093] (C2) In some embodiments of the method of (C1), the method further includes determining a dialogue context including the dialogue input and output. The method also includes retrieving and displaying supplemental content based on the dialogue context. Additionally, the method includes generating and presenting one or more selectable query suggestions with the generative language model, the selectable query suggestions based on the dialogue input and output.

[0094] (C3) In some embodiments of at least one of the methods of (C1)-(C2), the method further includes prompting the user to switch to SERP mode in response to a user selection of a selectable query suggestion among the one or more selectable query suggestions.

[0095] (C4) In some embodiments of at least one of methods (C1) through (C3), the method further includes receiving and displaying one or more selectable hyperlinks corresponding to one or more information sources included within the interactive output, the selectable hyperlinks being usable by a search engine when selected to search the one or more information sources.

[0096] (C5) In some embodiments of at least one of methods (C1) through (C4), the method further includes receiving an indication that the user has performed an additional interface mode change action and resuming display of interactive input and output.

[0097] (C6) In some embodiments of at least one of the methods (C1) through (C5), the interface mode change action includes at least one of swiping on a graphical user interface, selecting a selectable graphical icon, a voice command, and operating a slider bar to switch between a SERP mode and a conversation mode.

[0098] (D1) In another aspect, a method is described herein that is performed by a computing device, wherein the method includes any of the acts described in embodiments (A1) through (A7).

[0099] What has been described above includes examples of one or more embodiments. Of course, it is not possible to describe every conceivable modification and variation of the above-described devices or methodologies for the purposes of describing the above aspects, but one skilled in the art may recognize that many further modifications and permutations of the various aspects are possible. Accordingly, the described aspects are intended to embrace all such alternatives, modifications, and variations that fall within the spirit and scope of the appended claims. Furthermore, to the extent that the term "include" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprise" interpreted when employed as a transitional term in a claim.

Claims

1. a processor, and 1. A computing device including a memory storing instructions that, when executed by the processor, cause the processor to: generating a prompt to be input to a generative language model, the prompt including a conversational input set by a user; providing the prompt as an input to the generative language model; receiving a conversational output from the generative language model that generated the conversational output based on the prompt; receiving an indication that the user has performed an interface mode change action; updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model; presenting the updated SERP to the user on a client computing device; A computing device that causes an action to be performed, including

2. The computing device of claim 1 , wherein the actions further include determining a dialogue context including the conversational input and output, and receiving and displaying supplemental content based on the dialogue context.

3. 3. The computing device of claim 1 or 2, wherein the acts further include receiving and displaying from the generative language model one or more selectable query suggestions generated by the generative language model based on the conversational input and output.

4. 4. The computing device of claim 1, wherein the actions further include receiving and displaying one or more selectable hyperlinks corresponding to one or more information sources included within the interactive output, the hyperlinks, when selected, being usable by a search engine to search the one or more information sources.

5. 5. The computing device of claim 1, wherein the action further comprises receiving an indication that the user has performed an additional interface mode change action and resuming the display of the interactive input and output.

6. 6. The computing device of claim 1, wherein the interface mode change action includes at least one of swiping on a graphical user interface, selecting a selectable graphical icon, a voice command, and operating a slider bar to switch between a SERP mode and a conversation mode.

7. 1. A method that facilitates providing a dual-mode search interface on a computing device, comprising: receiving a prompt as input to a generative language model, the prompt including a conversational input set by a user; generating and displaying a conversational output from the generative language model, the conversational output being generated based on the prompt; and receiving an indication that the user has performed an interface mode change action; updating a search engine results page (SERP) to provide information related to the conversational output generated by the generative language model; providing the updated SERP to the user on a client computing device.

8. determining a dialogue context including said conversational input and output; retrieving and displaying supplemental content based on the dialogue context; generating and presenting one or more selectable query suggestions with the generative language model, the selectable query suggestions based on the conversational input and output; and The method of claim 7 further comprising:

9. The method of claim 7 or 8, further comprising prompting the user to switch to a SERP mode in response to a user selection of a selectable query suggestion among the one or more selectable query suggestions.

10. 10. The method of any one of claims 7 to 9, further comprising generating and displaying one or more selectable hyperlinks corresponding to one or more information sources included within the interactive output, the selectable hyperlinks, when selected, being usable by a search engine to search the one or more information sources.

11. The method of any one of claims 7 to 10, further comprising receiving an indication that the user has performed an additional interface mode changing action, and resuming the display of the interactive input and output.

12. 12. The method of claim 7, wherein the interface mode change action comprises at least one of swiping on a graphical user interface, selecting a selectable graphical icon, a voice command, and operating a slider bar to switch between a SERP mode and a conversation mode.

Images