Method for training search term generation model and method and device for generating search term
By incorporating relevance scoring and a combined loss function, the search term generation model addresses the issues of relevance and overexposure, enhancing the quality and relevance of generated search terms.
Patent Information
- Application Number
- CN202310651669.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-02
- Publication Date
- 2024-12-03
AI Technical Summary
Existing search term generation methods lack relevance calculation and are prone to overexposure, resulting in poor quality and repetitive search terms.
A method involving a pre-trained model, relevance scoring, classification of search terms, and a combined loss function using maximum likelihood and contrastive learning to train a search term generation model, ensuring high relevance and quality.
The trained model generates search terms with high correlation to the text, improving their relevance and quality significantly.
Smart Images

Figure CN119066146A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence, and in particular to a method for training a search word generation model, a method and device for generating search words, a storage medium, an electronic device, and a computer program product. Background Art
[0002] Search terms refer to the words used by users when searching. We hope to generate a series of search terms that are highly relevant to the text through the text, so as to provide support for subsequent search recall and text recommendation.
[0003] The existing search term generation methods have the following two main disadvantages:
[0004] 1. Since the current method is to directly input document content into the Transformer model for training and reasoning, the model only encodes the text and lacks the process of calculating the text and search terms. Therefore, many of the generated search terms are not relevant to the document.
[0005] 2. The task of generating search terms belongs to the field of text generation, so the generated text will encounter the problem of overexposure, resulting in poor quality of generated search terms and repeated words. The overexposure problem means that during training, the training data is remembered by the model, so during reasoning, when encountering training data that has not been seen before, the model cannot generate appropriate search terms.
[0006] Therefore, existing search term generation methods are not ideal. Summary of the invention
[0007] Embodiments of the present application provide a method for training a search term generation model, a method and device for generating search terms, a storage medium, an electronic device, and a computer program product.
[0008] In a first aspect, an embodiment of the present application provides a method for training a search term generation model for an electronic device, the method comprising:
[0009] The generation step is to input the text into the pre-trained model to generate multiple search terms;
[0010] a calculation step, inputting the plurality of search terms and the text into a relevance model, and calculating a relevance score between each of the search terms and the text;
[0011] a classification step, based on the plurality of relevance scores, classifying one or more first search terms among the plurality of search terms as positive samples, and classifying one or more second search terms among the plurality of search terms as negative samples;
[0012] The training step uses a first loss function composed of the multiple search words and a second loss function composed of the positive samples and the negative samples to train the pre-trained model to obtain a trained search word generation model.
[0013] In a possible implementation of the first aspect, in the classification step, the search term having a relevance score greater than a first threshold is used as the first search term, and the search term having a relevance score less than a second threshold is used as the second search term.
[0014] The first threshold is different from the second threshold.
[0015] In a possible implementation of the first aspect above, the first loss function is assigned a first weight, and the second loss function is assigned a second weight.
[0016] In a possible implementation of the first aspect above, the first weight and the second weight are dynamically adjusted during training the pre-trained model.
[0017] In a possible implementation of the first aspect above, the pre-trained model is a pre-trained MT5 model, and the correlation model is a BERT model.
[0018] In a possible implementation of the first aspect above, the first loss function is a maximum likelihood loss function, and the second loss function is a contrastive learning loss function. .
[0019] In a second aspect, an embodiment of the present application provides a method for generating a search term for an electronic device, the method comprising:
[0020] Get steps, get text;
[0021] The generation step is to input the text into the search term generation model trained according to the first aspect to generate multiple search terms.
[0022] In a third aspect, an embodiment of the present application provides a search method for an electronic device, the method comprising: searching using one or more of a plurality of search terms generated according to the second aspect.
[0023] In a fourth aspect, an embodiment of the present application provides a device for training a search term generation model, the device comprising:
[0024] The generation unit inputs the text into the pre-trained model to generate multiple search terms;
[0025] A calculation unit, inputting the plurality of search terms and the text into a relevance model, and calculating a relevance score between each of the search terms and the text;
[0026] a classification unit, which classifies one or more first search terms among the plurality of search terms as positive samples, and classifies one or more second search terms among the plurality of search terms as negative samples based on the plurality of relevance scores;
[0027] The training unit trains the pre-trained model using a first loss function composed of the multiple search terms and a second loss function composed of the positive samples and the negative samples to obtain a trained search term generation model.
[0028] In a fifth aspect, an embodiment of the present application provides a device for generating a search term, the device comprising:
[0029] Get unit, get text;
[0030] A generating unit inputs the text into the trained search term generation model according to the fourth aspect to generate a plurality of search terms.
[0031] In a sixth aspect, an embodiment of the present application provides a computer program product, comprising computer executable instructions, which are executed by a processor to implement the method of training a search term generation model described in the first aspect, the method of generating search terms described in the second aspect, or the search method described in the third aspect.
[0032] In the seventh aspect, an embodiment of the present application provides a computer-readable storage medium, on which instructions are stored, and when the instructions are executed on a computer, the computer executes the method for training a search term generation model described in the first aspect, the method for generating search terms described in the second aspect, or the search method described in the third aspect.
[0033] In an eighth aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; one or more memories; wherein the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes the method for training a search term generation model described in the first aspect, the method for generating search terms described in the second aspect, or the search method described in the third aspect.
[0034] In the process of training the search term generation model, the present invention adopts a correlation model to construct a contrastive learning loss function, and combines the contrastive learning loss function with the maximum likelihood loss function by multi-task learning. Therefore, the trained search term generation model in the present invention can ensure that the generated search terms maintain a high degree of correlation with the text, and also greatly improve the quality of the search terms. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1According to some embodiments of the present application, a block diagram of an electronic device is shown;
[0036] Figure 2 According to some embodiments of the present application, a flowchart of a method for training a search term generation model is shown;
[0037] Figure 3 According to some embodiments of the present application, a flow chart of a method for generating a search term is shown;
[0038] Figure 4 According to some embodiments of the present application, a structural diagram of an apparatus for training a search term generation model is shown;
[0039] Figure 5 According to some embodiments of the present application, a structural diagram of a device for generating search terms is shown. DETAILED DESCRIPTION
[0040] Illustrative embodiments of the present application include, but are not limited to, methods for training a search term generation model, methods, apparatuses, media, electronic devices, and computer program products for generating search terms.
[0041] The embodiments of the present application will be described in further detail below in conjunction with the accompanying drawings.
[0042] Figure 1 According to some embodiments of the present application, a block diagram of an electronic device is shown.
[0043] like Figure 1 As shown, the electronic device 100 may include one or more processors 102, a system motherboard 108 connected to at least one of the processors 102, a system memory 104 connected to the system motherboard 108, a non-volatile memory (NVM) 106 connected to the system motherboard 108, and a network interface 110 connected to the system motherboard 108.
[0044] The processor 102 may include one or more single-core or multi-core processors. The processor 102 may include any combination of a general-purpose processor (CPU) and a special-purpose processor (e.g., a graphics processor, an application processor, a baseband processor, etc.). A graphics processing unit (GPU) is a special-purpose processor that has a higher number of cores and powerful parallel computing capabilities than a general-purpose processor and is widely used in computer graphics processing. In an embodiment of the present invention, the processor 102 may be configured to execute the following: Figure 2 Or one or more of the various embodiments shown in 3.
[0045] In some embodiments, the system board 108 may include any suitable interface controller (not shown). Figure 1), to provide any suitable interface to at least one of the processors 102 and / or any suitable device or component in communication with the system motherboard 108.
[0046] In some embodiments, the system board 108 may include one or more memory controllers to provide an interface to the system memory 104. The system memory 104 may be used to load and store data and / or instructions 120. In some embodiments, the system memory 104 of the electronic device 100 may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).
[0047] The non-volatile memory 106 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions 120. In some embodiments, the non-volatile memory 106 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a HDD (Hard Disk Drive), a CD (Compact Disc) drive, and a DVD (Digital Versatile Disc) drive.
[0048] The non-volatile memory 106 may include a portion of storage resources installed on the device of the electronic device 100, or it may be accessible to an external device but is not necessarily a portion of the external device. For example, the non-volatile memory 106 may be accessed over a network via the network interface 110.
[0049] In particular, the system memory 104 and the non-volatile memory 106 may respectively include: a temporary copy and a permanent copy of the instructions 120. The instructions 120 may include: when executed by at least one of the processors 102, causing the electronic device 100 to implement the following Figure 3 In some embodiments, the instructions 120 , hardware, firmware, and / or software components thereof may additionally / alternatively be located in the system board 108 , the network interface 110 , and / or the processor 102 .
[0050] The network interface 110 may include a transceiver for providing a radio interface for the electronic device 100, thereby communicating with any other suitable device (e.g., a front-end module, an antenna, etc.) through one or more networks. In some embodiments, the network interface 110 may be integrated with other components of the electronic device 100. For example, the network interface 110 may be integrated with at least one of the processor 102, the system memory 104, the non-volatile memory 106, and a firmware device (not shown) having instructions. When at least one of the processors 102 executes the instructions, the electronic device 100 implements Figure 2Or one or more of the various embodiments shown in 3.
[0051] The network interface 110 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 110 may be a network adapter, a wireless network adapter, a telephone modem, and / or a wireless modem.
[0052] In one embodiment, at least one of the processors 102 may be packaged together with one or more controllers for the system motherboard 108 to form a system in package (SiP). In one embodiment, at least one of the processors 102 may be integrated on the same die with one or more controllers for the system motherboard 108 to form a system on a chip (SoC).
[0053] The electronic device 100 may further include an input / output (I / O) device 112 connected to the system motherboard 108. The I / O device 112 may include a user interface to enable a user to interact with the electronic device 100; the design of the peripheral component interface enables the peripheral components to interact with the electronic device 100. In some embodiments, the electronic device 100 further includes a sensor for determining at least one of an environmental condition and location information related to the electronic device 100.
[0054] In some embodiments, the I / O device 112 may include, but is not limited to, a display (e.g., a liquid crystal display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., a light emitting diode flash), a keyboard, and a graphics card. The graphics card is composed of a graphics processor integrated with an I / O interface (e.g., a PCIE interface) that complies with the data transmission protocol specification.
[0055] In some embodiments, the peripheral component interface may include, but is not limited to, a non-volatile memory port, an audio jack, and a power interface.
[0056] In some embodiments, the sensors may include, but are not limited to, gyroscope sensors, accelerometers, proximity sensors, ambient light sensors, and positioning units. The positioning unit may also be part of or interact with the network interface 110 to communicate with components of a positioning network (e.g., global positioning system (GPS) satellites).
[0057] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the electronic device 100. In other embodiments of the present application, the electronic device 100 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.
[0058] Program code can be applied to input instructions to perform the functions described in the present invention and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, a system for processing instructions including processor 102 includes any system having a processor such as a digital signal processor (DSP), a microcontroller, an application specific integrated circuit (ASIC), or a microprocessor.
[0059] Program code can be implemented with high-level programming language or object-oriented programming language to communicate with the processing system. When necessary, program code can also be implemented with assembly language or machine language. In fact, the mechanism described in the present invention is not limited to the scope of any specific programming language. In either case, the language can be a compiled language or an interpreted language.
[0060] One or more aspects of at least one embodiment may be implemented by instructions stored on a computer-readable storage medium, which, when read and executed by a processor, enables an electronic device to implement the method of the embodiment described in the present invention.
[0061] The method for training a search term generation model, the method for generating search terms, and the search method provided in this application can be applied to Figure 1 In the electronic device 100 shown, the electronic device 100 is, for example, a computer, a mobile device, a platform device, etc.
[0062] Figure 2 4 is a flowchart of a method for training a search term generation model according to an embodiment of the present invention.
[0063] In the generation step S21, the text is input into the pre-trained model to generate multiple search terms.
[0064] The pre-trained model is, for example, a pre-trained MT5 (massively multilingual pre-trained text-to-text transformer) model. MT5 decodes the content of an input document (note) and then generates multiple search terms. Compared with the Transformer model used in the prior art, the pre-trained MT5 model can decode the content of the text more effectively, thereby generating higher quality search terms.
[0065] In the present invention, in order to ensure that more search terms are generated, more search terms are generated by random sampling. The pre-trained MT5 model can generate, for example, 150 search terms for each text.
[0066] In calculation step S22, multiple search terms and texts are input into the relevance model, and the relevance score of each search term and text is calculated.
[0067] The relevance model is, for example, a BERT (Bidirectional Encoder Representations from Transformers) model. Specifically, the 150 search terms and the document generated in S21 are input into the BERT model, and the relevance scores of the 150 search terms and the document are calculated, thereby obtaining, for example, 150 relevance scores.
[0068] In the classification step S23, based on the multiple relevance scores, one or more first search terms among the multiple search terms are classified as positive samples, and one or more second search terms among the multiple search terms are classified as negative samples.
[0069] As described above, 150 relevance scores are obtained. Specifically, the search terms with relevance scores greater than a first threshold are used as first search terms, and the search terms with relevance scores less than a second threshold are used as second search terms.
[0070] The first threshold is, for example, 0.75, and search terms with a relevance score greater than 0.75 are taken as first search terms. These first search terms represent search terms with good relevance to the text and are classified as positive samples. The second threshold is, for example, 0.5, and search terms with a relevance score less than 0.5 are taken as second search terms. These second search terms represent search terms with poor relevance to the text and are classified as negative samples.
[0071] It will be appreciated that the first threshold may be different from the second threshold.
[0072] In the training step S24, the pre-trained model is trained using a first loss function consisting of multiple search terms and a second loss function consisting of positive samples and negative samples to obtain a trained search term generation model.
[0073] In the process of training a deep neural network, because we hope that the output of the deep neural network is as close as possible to the value we really want to predict, we can compare the predicted value of the current network with the target value we really want, and then update the weight vector of each layer of the neural network according to the difference between the two (of course, there is usually an initialization process before the first update, that is, pre-configuring parameters for each layer in the deep neural network); for example, if the predicted value of the network is high, adjust the weight vector to make it predict a lower value, and keep adjusting it until the deep neural network can predict the target value we really want or a value very close to the target value we really want. Therefore, it is necessary to pre-define "how to compare the difference between the predicted value and the target value", which is the loss function, which is an important equation for measuring the difference between the predicted value and the target value. Among them, the higher the output value (loss) of the loss function, the greater the difference, so the training of the deep neural network becomes a process of minimizing this loss as much as possible.
[0074] The loss function includes multiple loss functions. In this embodiment, the first loss function is the maximum likelihood loss function, and the second loss function is the contrastive learning loss function. In the present invention, learning using the maximum likelihood loss function and the contrastive learning loss function can also be called multi-task learning.
[0075] Specifically, the 150 search terms are used to form a maximum likelihood loss function L1, and the positive samples and negative samples are used to form a contrastive learning loss function L2. The contrastive learning loss function is, for example, an InfoNCE loss function.
[0076] The maximum likelihood loss function L1 is assigned a first weight w1, and the contrastive learning loss function L2 is assigned a second weight w2.
[0077] The total loss function L of the pre-trained MT5 model can be expressed as L=w1*L1+w2*L2. This total loss function L is used to calculate the loss and perform stochastic gradient descent learning to train the pre-trained MT5 model and obtain the trained MT5 model, which is used as a search term generation model.
[0078] During the training of the pre-trained MT5 model, the first weight w1 and the second weight w2 may be adjusted dynamically.
[0079] Specifically, the pre-trained MT5 model can randomly select the values of w1 and w2, and perform stochastic gradient descent learning, taking the w1 and w2 with the minimum loss as the optimal w1 and w2.
[0080] Figure 3 is a flowchart of a method for generating search terms according to an embodiment of the present invention.
[0081] like Figure 3 As shown, in the acquisition step S31, the current text is acquired.
[0082] In the generation step S32, the text input Figure 2 The trained search term generation model shown, i.e., the trained MT5 model, generates a plurality of search terms.
[0083] The present invention also provides a search method, the method comprising: using Figure 3 One or more of the multiple search terms generated are searched.
[0084] In the process of training the search term generation model, the present invention adopts a correlation model to construct a contrastive learning loss function, and combines the contrastive learning loss function with the maximum likelihood loss function by multi-task learning. Therefore, the trained search term generation model in the present invention can ensure that the generated search terms maintain a high degree of correlation with the text, and also greatly improve the quality of the search terms.
[0085] The present invention also provides a device 40 for training a search term generation model, such as Figure 4 As shown, the device 40 includes: a generation unit 41, inputting text into a pre-trained model to generate multiple search terms; a calculation unit 42, inputting the multiple search terms and the text into a correlation model, and calculating the correlation score between each search term and the text; a classification unit 43, based on the multiple correlation scores, classifying one or more first search terms among the multiple search terms as positive samples, and classifying one or more second search terms among the multiple search terms as negative samples; a training unit 44, training the pre-trained model using a first loss function composed of the multiple search terms and a second loss function composed of the positive samples and the negative samples to obtain a trained search term generation model.
[0086] It is understandable that the generation unit 41, the calculation unit 42, the classification unit 43, and the training unit 44 can be implemented by the processor 102 having the functions of these modules or units in the electronic device 100. The implementation methods disclosed in the foregoing are method implementation methods corresponding to this implementation method, and this implementation method can be implemented in conjunction with the above-mentioned implementation methods. The relevant technical details mentioned in the above-mentioned implementation methods are still valid in this implementation method, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this implementation method can also be applied in the above-mentioned implementation methods.
[0087] The present invention also provides a device 50 for generating search terms, such as Figure 5 As shown, the device 50 includes: an acquisition unit 51, which acquires text; a generation unit 52, which inputs the text into Figure 4 The device 50 generates a trained search term generation model to generate multiple search terms.
[0088] It is understandable that the acquisition unit 51 and the generation unit 52 can be implemented by the processor 102 having the functions of these modules or units in the electronic device 100. The embodiments disclosed in the foregoing are method embodiments corresponding to this embodiment, and this embodiment can be implemented in conjunction with the above-mentioned embodiments. The relevant technical details mentioned in the above-mentioned embodiments are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above-mentioned embodiments.
[0089] The present invention also provides a computer program product, including computer executable instructions, which are executed by the processor 102 to implement the method of training a search term generation model or the method of generating search terms or the search method of the present invention.
[0090] The present invention also provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed on a computer, the computer executes the method for training a search term generation model or the method for generating search terms or the search method of the present invention.
[0091] It should be noted that in the examples and description of this patent, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one" do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0092] Although the present application has been illustrated and described with reference to certain preferred embodiments thereof, it will be apparent to those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present application.
[0093] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0094] It should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the intention that the claimed invention requires more features than those expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in less than all of the features of the individual embodiments previously disclosed. Therefore, the claims that follow the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.
[0095] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.
[0096] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.
Claims
1. A method for training a search term generation model for an electronic device, characterized in that: The method comprises: The generation step inputs the text into the pre-trained model to generate multiple search terms; a calculation step of inputting the plurality of search terms and the text into a relevance model to calculate a relevance score between each search term and the text; a classification step of classifying one or more first search terms among the plurality of search terms as positive samples and classifying one or more second search terms among the plurality of search terms as negative samples based on the plurality of relevance scores; The training step uses a first loss function composed of the multiple search terms and a second loss function composed of the positive samples and the negative samples to train the pre-trained model to obtain a trained search term generation model.
2. The method according to claim 1, characterized in that In the classification step, the search terms with a relevance score greater than a first threshold are used as the first search terms, and the search terms with a relevance score less than a second threshold are used as the second search terms, The first threshold is different from the second threshold.
3. The method according to claim 1, characterized in that The first loss function is assigned a first weight, and the second loss function is assigned a second weight.
4. The method according to claim 3, characterized in that During the training of the pre-trained model, the first weight and the second weight are dynamically adjusted.
5. The method according to any one of claims 1 to 4, characterized in that The pre-trained model is a pre-trained MT5 model, and the correlation model is a BERT model.
6. The method according to any one of claims 1 to 4, characterized in that The first loss function is a maximum likelihood loss function, and the second loss function is a contrastive learning loss function.
7. A method for generating a search term, for use in an electronic device, characterized in that: The method comprises: Get steps, get text; The generation step is to input the text into the search term generation model trained according to any one of claims 1-6 to generate multiple search terms.
8. A search method for an electronic device, characterized in that: The method comprises searching using one or more of a plurality of search terms generated according to claim 7 .
9. A device for training a search term generation model, characterized in that: The device comprises: The generation unit inputs the text into the pre-trained model and generates multiple search terms; a calculation unit, inputting the plurality of search terms and the text into a relevance model, and calculating a relevance score between each search term and the text; a classification unit, which classifies one or more first search terms among the plurality of search terms as positive samples, and classifies one or more second search terms among the plurality of search terms as negative samples based on the plurality of relevance scores; The training unit trains the pre-training model using a first loss function composed of the multiple search terms and a second loss function composed of the positive samples and the negative samples to obtain a trained search term generation model.
10. A device for generating a search term, characterized in that: The device comprises: Get cell, get text; A generating unit inputs the text into the trained search term generation model according to claim 10 to generate a plurality of search terms.
11. A computer program product comprising computer executable instructions, characterized in that: The instructions are executed by a processor to implement the method for training a search term generation model according to any one of claims 1 to 6, the method for generating search terms according to claim 7, or the search method according to claim 8.
12. A computer-readable storage medium, characterized in that The storage medium stores instructions, which, when executed on a computer, enable the computer to execute the method for training a search term generation model according to any one of claims 1 to 6, the method for generating search terms according to claim 7, or the search method according to claim 8.
13. An electronic device, characterized in that: include: one or more processors; one or more memories; Wherein, the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes the method for training a search term generation model as described in any one of claims 1 to 6, the method for generating search terms as described in claim 7, or the search method as described in claim 8.
Citation Information
Cited By
Search keyword generation model training method
CN122412963A
