Workflow generation method and device, equipment, medium and product

Abstract

The invention discloses a workflow generation method, device and equipment, a medium and a product. The method relates to the technical field of computers, and comprises the following steps: displaying a studio editing interface in response to a preset trigger operation; in response to a node selection operation of a user on the studio editing interface, calling preset function nodes, the function nodes including at least one of a large language model node, a retrieval node, a start node, an end node, a logic node, a conversion node and a tool node; in response to a configuration operation of a user for each node on the studio editing interface, configuration information of each function node is acquired, and the configuration information of each function node comprises node attribute information and a node relationship; the workflow is generated based on the node attribute information and the node relationship of each function node on the workroom editing interface, and through the technical scheme of the invention, the workflow which can deeply cooperate with a large model, is efficient, adapts to multiple service scenes and has a low threshold can be generated.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a workflow generation method, apparatus, device, medium and product. Background Technology

[0002] With the rapid development of artificial intelligence technology, large-scale language models, with their powerful semantic understanding and generation capabilities, have been widely applied in various business scenarios such as customer service, semantic search, and office collaboration.

[0003] In the practical application of large-scale language models, several bottlenecks still need to be overcome, mainly manifested in three core issues: high implementation threshold, high deployment and maintenance costs, and long project cycles. Specifically, on the one hand, existing large-scale model applications are mostly limited to single question-and-answer or simple generation scenarios, lacking the ability to adapt to complex business processes and unable to connect multiple tasks, tools, and data sources through standardized processes. On the other hand, in complex business scenarios, relying solely on the native capabilities of large-scale models is easily constrained by the quality of prompt words and the limits of inference, resulting in insufficient task execution accuracy, poor process controllability, and high costs for manual intervention, making it difficult to achieve automated and large-scale implementation. Against this backdrop, how to build a workflow that can deeply collaborate with large-scale models, is efficient, adaptable to multiple business scenarios, and has a low threshold for implementation has become a key technical problem that urgently needs to be solved in the current process of implementing artificial intelligence technology. Summary of the Invention

[0004] This invention provides a workflow generation method, apparatus, device, medium, and product that can build workflows that are highly collaborative with large models, efficient, adaptable to multiple business scenarios, and easy to implement.

[0005] According to one aspect of the present invention, a workflow generation method is provided, comprising: In response to a preset trigger action, the studio editing interface is displayed; In response to a user's node selection operation on the studio editing interface, a pre-set functional node is invoked, including at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node; In response to the user's configuration operations on each node in the studio editing interface, the configuration information of each functional node is obtained, wherein the configuration information of each functional node includes: node attribute information and node relationship; A workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface.

[0006] According to another aspect of the present invention, a workflow generation apparatus is provided, the workflow generation apparatus comprising: The display module is used to display the studio editing interface in response to preset trigger operations; The calling module is used to call pre-set function nodes in response to the user's node selection operation on the studio editing interface. The function nodes include at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node. The acquisition module is used to acquire the configuration information of each functional node in response to the user's configuration operation on each node in the studio editing interface. The configuration information of each functional node includes: node attribute information and node relationship. The generation module is used to generate workflows based on the node attribute information and node relationships of each functional node on the studio editing interface.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the workflow generation method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the workflow generation method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer program product is provided, which, when executed by a processor, implements the workflow generation method as described in any of the embodiments of the present invention.

[0010] This invention, in response to a preset trigger operation, displays a studio editing interface; in response to a user's node selection operation on the studio editing interface, it invokes pre-set functional nodes, including at least one of the following: large language model nodes, retrieval nodes, start nodes, end nodes, logic nodes, transformation nodes, and tool nodes; in response to a user's configuration operation for each node on the studio editing interface, it obtains the configuration information of each functional node, including node attribute information and node relationships; and generates a workflow based on the node attribute information and node relationships of each functional node on the studio editing interface, enabling the generation of a workflow that can deeply collaborate with large models, is efficient, adaptable to multiple business scenarios, and has a low barrier to entry for implementation.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a workflow generation method in an embodiment of the present invention; Figure 2 This is a schematic diagram of a PanaHub application architecture in an embodiment of the present invention; Figure 3 This is a principle block diagram of an embodiment of the present invention; Figure 4 This is an overall flowchart of PanaHub in one embodiment of the present invention; Figure 5 This is a schematic diagram of a workflow generation process in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a workflow generation device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0017] Example 1 Figure 1 This is a flowchart illustrating a workflow generation method provided in an embodiment of the present invention. This embodiment is applicable to workflow generation scenarios. The method can be executed by a workflow generation device in this embodiment, which can be implemented in software and / or hardware, such as... Figure 1 As shown, the method specifically includes the following steps: S110, in response to a preset trigger operation, displays the studio editing interface.

[0018] In this embodiment, the large model invocation tool (which can be named PanaHub) connecting terminal devices operates within an independently controllable system, primarily combining AI large language models and Retrieval-Augmented Generation (RAG) technology. Combined with hyperconverged cloud products, it provides users with a platform that can orchestrate intelligent agent workflows according to business needs, enabling functions such as intelligent customer service, intelligent tools, and intelligent database management. This helps enterprises reduce development costs and improves employee work efficiency and reduces work difficulty.

[0019] PanaHub application architecture as follows Figure 2As shown, the basic resource layer is a hyperconverged cloud service platform, which connects to PanaHub as a plugin for flexible user access. This generates PanaHub interfaces and a management layer for managing application services such as workflows. The studio is used to create workflows and chatflow windows. The knowledge base serves as a repository for model inference and training, and the tools section provides commonly used tools. The core functional area manages specific data such as session count, user count, and token output speed through real-time performance and status monitoring. Large model application lifecycle management provides interface services as a unified upper-level management backend. The AI ​​application layer mainly consists of large model inference applications and enterprise-customized AI applications, and can integrate large models such as GPT-4 and Tongyi, performing overall framework computation based on OpenAI's large model algorithms. The technical solution provided in this embodiment offers a unified calling interface, shielding the differences in APIs between different models. Furthermore, it can dynamically allocate container resources based on model performance prediction results, ensuring a match between model inference load and resource allocation, avoiding resource idleness or overload (container resources are provided by the hyperconverged infrastructure).

[0020] In this embodiment, the principle block diagram is as follows: Figure 3 As shown, the application layer provides a visual editing page, workflow node library, workflow canvas, application API interface, and real-time preview and debugging. The core engine layer includes a workflow parsing module, agent framework, model scheduling module, and RAG engine. The model service layer includes a resource management module and a model API gateway (online large models [GPT-4 / Claude / deepseek, etc.], local large model services [Llama / Mistral / Qwen]). The data storage layer uses Redis to cache intermediate states and PostgreSQL data service management, providing vectorized data storage, dialogue history storage, application configuration storage, model configuration storage, and structured data storage.

[0021] In this embodiment, the preset trigger operation can be a trigger operation on the target application icon, or it can be other operations that trigger the studio editing interface, such as mouse click operations, keyboard input operations, etc. For example, the terminal device may display the studio editing interface on the terminal device's screen in response to preset trigger operations such as clicking a virtual button or pressing a physical button.

[0022] In this embodiment, the studio editing interface is a visual operation interface, which refers to the operation interface for building workflows by dragging and dropping nodes.

[0023] S120, in response to the user's node selection operation on the studio editing interface, calls the pre-set function node.

[0024] In this embodiment, the functional nodes include at least one of the following: large language model node, retrieval node, start node, end node, logic node, transformation node, and tool node.

[0025] In this embodiment, in response to a user's node selection operation on the studio editing interface, the way to call a pre-set function node can be: in response to a user's trigger operation in the display area, display a node list; in response to a user's selection operation of any node in the node list, call the corresponding function node and display the node in the display area.

[0026] Optionally, the studio editing interface includes a display area and an editing area; In response to a user's node selection operation on the studio editing interface, pre-set function nodes are invoked, including: In response to a user's triggered action in the display area, a list of nodes is displayed.

[0027] In this embodiment, the node list includes at least one of the following: large language model nodes, retrieval nodes, start nodes, end nodes, logic nodes, transformation nodes, and tool nodes. For example, the node list may include: LLM nodes, knowledge retrieval nodes, direct reply nodes, agent nodes, question classifier nodes, start nodes, logic nodes, transformation nodes, and tool nodes. The logic nodes include: conditional branch nodes, iteration nodes, and loop nodes. The transformation nodes include: code execution nodes, template transformation nodes, variable aggregator nodes, document extractor nodes, variable assignment nodes, and parameter extractor nodes. The tool nodes include: HTTP request nodes and list operation nodes. In this embodiment, the tool nodes may also include: plug-in tool nodes, custom tool nodes, and workflow tool nodes. In a specific example, the tool nodes include: audio / video nodes, visualization chart nodes, test plug-in nodes, code interpreter nodes, Excel toolbox nodes, time nodes, data analysis nodes, web scraping nodes, etc.

[0028] In response to a user's selection of any node in the node list, the corresponding functional node is invoked and displayed in the display area.

[0029] In a specific example, in response to the user's selection of a start node in the node list, the corresponding function node is invoked, and the start node is displayed in the display area.

[0030] S130, in response to the user's configuration operations on each node in the studio editing interface, obtains the configuration information of each functional node.

[0031] The configuration information for each functional node includes: node attribute information and node relationships.

[0032] In this embodiment, the node attribute information includes: relevant parameters of the functional node. For example, if the node is the start node, then the node attribute information of the functional node corresponding to the start node includes: In this embodiment, in response to the user's configuration operation on each node in the studio editing interface, the configuration information of each functional node can be obtained as follows: when a parameter editing operation of a node is detected, the node's attribute configuration interface is displayed, and the node attribute information of the corresponding functional node is obtained based on the attribute configuration interface; the connection lines set between each node are obtained, and the node relationship between the connected functional nodes is determined based on the connection lines.

[0033] Optionally, in response to user configuration operations on each node in the studio editing interface, the configuration information of each functional node is obtained, including: When a parameter editing operation is detected on a node, the node's attribute configuration interface is displayed, and the node attribute information of the corresponding functional node is obtained based on the attribute configuration interface.

[0034] In this embodiment, the parameter editing operation of a node can be a touch operation on a node in the studio editing interface.

[0035] In this embodiment, the node attribute configuration interface includes: an area displaying added variables and an add variable control. In response to a triggered operation on the add variable control in the node attribute configuration interface, the add variable interface is displayed. In response to a user's configuration operation on the add variable interface, the attribute information of the newly added variable is obtained. Based on the attribute information of the added variables in the area displaying added variables and the attribute information of the newly added variable, the node attribute information of the functional node is determined.

[0036] In a specific example, if the node is the start node, the add variable interface includes: an editing area, a cancel control, and a save control; the editing area includes: a field type selection area, a variable name input area, a display name input area, a file type selection area, and a maximum length input area; in response to detecting an input operation on the editing area of ​​the add variable interface, and after detecting a trigger operation on the save control of the add variable interface, the variable is added to the node's attribute configuration interface based on the parameters of the input editing area.

[0037] In another specific example, if the node is a start node, its attribute configuration interface includes parameters such as: field type (text, paragraph, dropdown option, number, single file, file list), variable name, display name, supported file types (document, image, audio, video, other file types), and uploaded file type (local upload, URL, both). If the node is a knowledge retrieval node, its attribute configuration interface includes: query variable input area, knowledge base selection area, and output variable input area. If the node is a document extractor node, its attribute configuration interface includes: input variable input area and output variable input area. If the node is an LLM node, its attribute configuration interface includes: model selection area and prompt word input area. If the node is a branch node, its attribute configuration interface includes: IF execution condition addition area. When the IF condition is not met, the corresponding ELSE branch is executed. If the node is a direct reply node, its attribute configuration interface includes: reply content input area. If the node is a variable aggregator node, its attribute configuration interface includes: variable assignment area and mixed grouping setting area. If the node is a code execution node, the attribute configuration interface of the code execution node includes: input variable editing area, code input area, and output variable editing area.

[0038] Obtain the connection lines set between each node, and determine the node relationship between the connected functional nodes based on the connection lines.

[0039] In this embodiment, the node relationship includes the previous node and the next node of the current node.

[0040] In this embodiment, the connection lines set between each node on the studio editing interface are obtained, and the node relationship between the connected functional nodes is determined based on the connection lines.

[0041] It should be noted that the area near the node on the studio editing interface displays associated controls, which are used to perform at least one of the following functions: running the node, changing the node, copying, deleting, and duplicating.

[0042] S140, a workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface.

[0043] In this embodiment, the workflow generated based on the node attribute information and node relationships of each functional node on the studio editing interface can be as follows: in response to the user's triggering operation of the preview control on the studio editing interface, a workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface.

[0044] Through the technical solutions provided by the embodiments of the present invention, users can better manage their business, output product-related questions through dialogue, help them understand relevant products more quickly, and improve service quality and efficiency.

[0045] The technical solution provided in this embodiment can break down complex tasks into smaller steps (nodes), reduce system complexity, reduce reliance on prompt word technology and model reasoning ability, and improve the performance of large language models in complex tasks.

[0046] Optionally, a workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface, including: In response to the user's triggering operation of the preview control on the studio editing interface, a workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface, the workflow is run, and a preview interaction window is displayed.

[0047] In this embodiment, the preview interaction window includes a historical dialogue display area and a target question text input area. The historical dialogue display area is used to display historical question texts and their corresponding response texts. The target question text input area is used to input the question to be answered.

[0048] Optionally, the current user's icon and system icon can also be displayed in the preview window.

[0049] In response to the user's editing actions in the preview interaction window, obtain the target question text.

[0050] In this embodiment, the target question text input by the user is obtained in response to the user's editing operation on the target question text input area.

[0051] It should be noted that after obtaining the target issue text, the target issue text is displayed in the preview interaction window.

[0052] Based on the target question text and the workflow, determine the response text corresponding to the target question text.

[0053] In this embodiment, the workflow is run based on the target question text to obtain the response text corresponding to the target text.

[0054] The preview window displays the response text corresponding to the target question text.

[0055] In this embodiment, the response text corresponding to the target question text is displayed below the display area of ​​the target question text.

[0056] Optional, also includes: In response to a user's triggering action on the publish control in the studio editing interface, the workflow on the studio editing interface is published.

[0057] It should be noted that after the workflow is published, it can be used by third parties.

[0058] Optional, also includes: In response to the user's triggering action on the knowledge base control in the studio editing interface, the user is redirected to the knowledge base settings interface.

[0059] In this embodiment, the knowledge base settings interface includes: a visibility permission selection area, an index mode setting area, a model type selection area, and a search settings area.

[0060] In response to a user's settings operation on the knowledge base settings interface, at least one of the knowledge base's indexing mode, model type, and retrieval settings is obtained.

[0061] In this embodiment, the indexing modes include: high-quality and economical. High-quality refers to calling the embedding model to process documents for more accurate retrieval, which can help large language models generate high-quality answers. Economical refers to using 10 keywords per block for retrieval, which does not consume tokens, but reduces retrieval accuracy.

[0062] In this embodiment, the retrieval settings include: vector retrieval, full-text retrieval, hybrid retrieval, weight settings, and a re-ranking model. Vector retrieval is used to generate query embeddings and query text segments most similar to their vector representations; full-text retrieval is used to index all words in the documents, allowing users to query any words and returning text fragments containing those words. Hybrid retrieval is used to perform full-text retrieval and vector retrieval simultaneously, and apply a re-ranking step to select the best result matching the user's question from similar query results. Users can choose to set weights or configure a re-ranking model. Weight settings are used to determine whether semantic matching or keyword matching is prioritized by adjusting the assigned weights in the re-ranking strategy. The re-ranking model is used to re-rank the candidate document list based on its semantic match with the user's question, thereby improving the semantic ranking results.

[0063] The technical solution provided in this embodiment can be combined with hyperconverged software to improve the overall office efficiency of users and is closely integrated with large models.

[0064] In a specific example, the overall flowchart of PanaHub is as follows: Figure 4As shown, based on the hyperconverged cloud platform, users can operate the PanaHub management page to manage studios, tools, and knowledge bases. Adding corresponding knowledge content to the knowledge base allows large models to have more comprehensive data reserves, making the content generated by the workflow more complete. Different knowledge base information can be generated according to different business processes, and then the information can be displayed in the form of dialogues through the workflow, making the output content more complete and the interaction more precise. In this embodiment, by combining the hyperconverged cloud product with PanaHub, complex AI tasks are broken down into visual workflow nodes, enabling enterprise applications to execute business processes more accurately, rather than simply interacting with a robot. It should be noted that the hyperconverged cloud platform is more lightweight, convenient, and adaptable compared to other cloud products.

[0065] In addition, after publishing the workflow as a tool, PanaHub data can also be output and stored on the hyperconverged cloud platform.

[0066] In another specific example, a flowchart illustrating the workflow generation process is as follows: Figure 5 As shown, users can click the Workflow tab in the studio, then click Create Blank Application or Create from Template, select the starting node, execute the required node information, and if there is a document uploaded, perform LLM node parsing. Then, by selecting other nodes and conditional nodes, and executing branches to the knowledge retrieval node, complete workflow information is generated. The workflow can be used as a template and referenced when setting the chatflow dialog box.

[0067] The technical solution of this embodiment displays a studio editing interface in response to a preset trigger operation; in response to a user's node selection operation on the studio editing interface, it calls pre-set functional nodes, including at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node; in response to a user's configuration operation for each node on the studio editing interface, it obtains the configuration information of each functional node, including node attribute information and node relationships; and generates a workflow based on the node attribute information and node relationships of each functional node on the studio editing interface. Through the technical solution of this invention, a workflow that can deeply collaborate with large models, is efficient, adaptable to multiple business scenarios, and has a low threshold for implementation can be generated.

[0068] Example 2 Figure 6 This is a schematic diagram of a workflow generation device provided in an embodiment of the present invention. This embodiment is applicable to workflow generation scenarios. The device can be implemented using software and / or hardware, and can be integrated into any device that provides workflow generation functionality, such as… Figure 6As shown, the workflow generation device specifically includes: a display module 610, a calling module 620, an acquisition module 630, and a generation module 640.

[0069] The display module is used to display the studio editing interface in response to preset trigger operations; The calling module is used to call pre-set function nodes in response to the user's node selection operation on the studio editing interface. The function nodes include at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node. The acquisition module is used to acquire the configuration information of each functional node in response to the user's configuration operation on each node in the studio editing interface. The configuration information of each functional node includes: node attribute information and node relationship. The generation module is used to generate workflows based on the node attribute information and node relationships of each functional node on the studio editing interface.

[0070] The above-described products can perform the methods provided in any embodiment of the present invention, and have the corresponding functional modules and beneficial effects for performing the methods.

[0071] Example 3 Figure 7 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0072] like Figure 7As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0073] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0074] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as workflow generation methods.

[0075] In some embodiments, the workflow generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the workflow generation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the workflow generation method by any other suitable means (e.g., by means of firmware).

[0076] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0077] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0078] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0079] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0080] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0081] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0082] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0083] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the workflow generation method according to any embodiment of the invention.

[0084] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0085] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A workflow generation method, characterized in that, include: In response to a preset trigger action, the studio editing interface is displayed; In response to a user's node selection operation on the studio editing interface, a pre-set functional node is invoked, including at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node; In response to the user's configuration operations on each node in the studio editing interface, the configuration information of each functional node is obtained, wherein the configuration information of each functional node includes: node attribute information and node relationship; A workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface.

2. The method according to claim 1, characterized in that, The studio editing interface includes a display area and an editing area; In response to a user's node selection operation on the studio editing interface, pre-set function nodes are invoked, including: In response to a user's trigger action in the display area, a list of nodes is displayed; In response to a user's selection of any node in the node list, the corresponding functional node is invoked and displayed in the display area.

3. The method according to claim 1, characterized in that, In response to user configuration operations on each node in the studio editing interface, the configuration information of each functional node is obtained, including: When a parameter editing operation is detected on a node, the node's attribute configuration interface is displayed, and the node attribute information of the corresponding functional node is obtained based on the attribute configuration interface. Obtain the connection lines set between each node, and determine the node relationship between the connected functional nodes based on the connection lines.

4. The method according to claim 1, characterized in that, A workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface, including: In response to the user's triggering operation of the preview control on the studio editing interface, a workflow is generated based on the node attribute information and node relationships of each functional node on the studio editing interface, the workflow is run, and a preview interaction window is displayed; In response to the user's editing actions in the preview interaction window, obtain the target question text; Based on the target question text and the workflow, determine the response text corresponding to the target question text; The preview window displays the response text corresponding to the target question text.

5. The method according to claim 4, characterized in that, Also includes: In response to a user's triggering action on the publish control in the studio editing interface, the workflow on the studio editing interface is published.

6. The method according to claim 1, characterized in that, Also includes: In response to the user's triggering operation on the knowledge base control in the studio editing interface, the user is redirected to the knowledge base settings interface; In response to a user's settings operation on the knowledge base settings interface, at least one of the knowledge base's indexing mode, model type, and retrieval settings is obtained.

7. A workflow generation device, characterized in that, include: The display module is used to display the studio editing interface in response to preset trigger operations; The calling module is used to call pre-set function nodes in response to the user's node selection operation on the studio editing interface. The function nodes include at least one of the following: large language model node, retrieval node, start node, end node, logic node, conversion node, and tool node. The acquisition module is used to acquire the configuration information of each functional node in response to the user's configuration operation on each node in the studio editing interface. The configuration information of each functional node includes: node attribute information and node relationship. The generation module is used to generate workflows based on the node attribute information and node relationships of each functional node on the studio editing interface.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the workflow generation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the workflow generation method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the workflow generation method according to any one of claims 1-6.

Images