Use of external knowledge

Abstract

Example embodiments of the present disclosure relate to the use of external knowledge. A method at a first device includes transmitting, to a second device, information related to use of external knowledge for a communication network service, where the external knowledge is stored outside a communication network function, and the communication network function is configured to operate the communication network service; sending a request for the communication network service to a second device; and in response to sending the request, receiving a response to the communication network service from the second device, where the response is based on at least one of internal knowledge of the communication network function, and external knowledge.

Description

TECHNICAL FIELD

[0001] Various example embodiments of the present disclosure generally relate to the field of telecommunications, and in particular to an apparatus, a method, and a computer-readable storage medium for use of external knowledge. BACKGROUND

[0002] Generative artificial intelligence (AI) is a method of creating different types of content following the characteristics of training data. One application of generative AI is the use of large language models (LLMs) to generate plausible text or language based on an input query. There are many examples of proprietary and open-source models currently available and used in different applications. A general-purpose LLM (as an example base model) can be obtained by extensive training using vast amounts of data in order to capture the relationships between words and obtain general-purpose capabilities for text understanding, processing, and generation. Such a process is called pre-training. Fine-tuning is the process of adapting a general-purpose model to a specific domain task, such as understanding technical text and recommending management actions in the telecommunications domain. This can be done by selectively adjusting or training a subset of model parameters or a set of additional parameters. SUMMARY

[0003] In a first aspect of the present disclosure, a first apparatus is provided. The first apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus at least to: send, to a second apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; send, to the second apparatus, a request for the communication network service; and in response to sending the request, receive, from the second apparatus, a response for the communication network service, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0004] In a second aspect of the present disclosure, a second apparatus is provided. The second apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus at least to: receive, from a first apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; receive, from the first apparatus, a request for the communication network service; and in response to receiving the request, send, to the first apparatus, a response for the communication network service, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0005] In a third aspect of the disclosure, a method is provided. The method comprises: sending, to a second apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; sending, to the second apparatus, a request for the communication network service; and receiving, from the second apparatus, a response for the communication network service in response to sending the request, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0006] In a fourth aspect of the disclosure, a method is provided. The method comprises: receiving, from a first apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; receiving, from the first apparatus, a request for the communication network service; and sending, to the first apparatus, a response for the communication network service in response to receiving the request, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0007] In a fifth aspect of the disclosure, a first apparatus is provided. The first apparatus comprises: means for sending, to a second apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; means for sending, to the second apparatus, a request for the communication network service; and means for receiving, from the second apparatus, a response for the communication network service in response to sending the request, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0008] In a sixth aspect of the disclosure, a second apparatus is provided. The second apparatus comprises: means for receiving, from a first apparatus, information related to use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; means for receiving, from the first apparatus, a request for the communication network service; and means for sending, to the first apparatus, a response for the communication network service in response to receiving the request, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0009] In a seventh aspect of the disclosure, a computer readable medium is provided. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the third aspect.

[0010] In an eighth aspect of the disclosure, a computer readable medium is provided. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the fourth aspect.

[0011] It is to be understood that the Summary is not intended to identify key or essential features of embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features, details, and advantages of the disclosure will become BRIEF DESCRIPTION OF DRAWINGS

[0012] Some example embodiments will now be described with reference to the accompanying drawings, in which:

[0013] Figure 1 An example communication environment in which example embodiments of the disclosure can be implemented is shown;

[0014] Figure 2 An example process for Retrieval Augmentation Generation (RAG) is shown, in accordance with some example embodiments of the disclosure;

[0015] Figure 3 A signaling diagram showing an example process for configuring the use of external knowledge, in accordance with some example embodiments of the disclosure;

[0016] Figure 4 A signaling diagram showing an example process for RAG configuration and reporting, in accordance with some example embodiments of the disclosure;

[0017] Figure 5A And Figure 5B An example use case regarding intent-based management is shown, in accordance with some example embodiments of the disclosure;

[0018] Figure 6A And Figure 6B An example use case for performing user device-specific tasks is shown, in accordance with some example embodiments of the disclosure;

[0019] Figure 7 A flow diagram showing an example method implemented at a first device, in accordance with some example embodiments of the disclosure;

[0020] Figure 8 A flow diagram showing an example method implemented at a second device, in accordance with some example embodiments of the disclosure;

[0021] Figure 9 is a simplified block diagram of a device suitable for implementing example embodiments of the disclosure; and

[0022] Figure 10 A block diagram showing an example computer-readable medium, in accordance with some example embodiments of the disclosure.

[0023] Throughout the drawings, identical or similar reference numerals can designate identical or similar elements. DETAILED DESCRIPTION

[0024] The principles of the present disclosure will now be described with reference to some example embodiments. It should be understood that these embodiments are described for illustrative purposes only and help the skilled person understand and implement the present disclosure without implying any limitation to the scope of the present disclosure. The embodiments described herein can be implemented in various ways other than those described below.

[0025] In the following description and claims, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

[0026] References in the present disclosure to “one embodiment”, “an embodiment”, “example embodiments”, etc., indicate that the embodiment described can include a particular feature, structure, or characteristic, but every embodiment can not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Furthermore, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of those skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0027] It should be understood that although the terms “first”, “second”, etc. can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element without departing from the scope of the example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed terms.

[0028] As used herein, “at least one of ” and “one or more of ” and the like means at least one, or, alternatively, at least one of any one of the listed elements.

[0029] As used herein, unless expressly stated otherwise, performing a step “in response to A” does not indicate that the step is performed immediately following the occurrence of “A”, but can include one or more intervening steps.

[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprise," "comprising," "have," "having," "include," and / or "including," when used herein, specify the presence of stated features, elements, and / or components, but do not preclude the presence or addition of one or more other features, elements, components, and / or groups thereof.

[0031] As used in this application, the term "circuitry" can refer to one or more or all of the following: (a) hardware-only circuitry such as amongst other things analog and / or digital circuitry; (b) combinations of hardware circuits and software, such as (as applicable): (i) combinations of analog and / or digital hardware circuit(s) with software / firmware; (ii) portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions and (c) hardware circuit(s) and / or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software can not be present when it is not needed for operation.

[0032] This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation that includes one or more processors and / or processor cores and / or a portion of a processor and / or processor core and their accompanying software and / or firmware. The term circuitry also covers, for example, if applicable to a particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in a server, a cellular network device, or other computing or network device.

[0033] As used herein, the term “communication network” refers to a network that follows any suitable communication standard, such as New Radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed Packet Access (HSPA), Narrow Band Internet of Things (NB-IoT), etc. Also, communication between terminal devices and network devices in a communication network can be performed according to any suitable generation communication protocol that is currently known or will be developed in the future. Embodiments of the present disclosure can be applied to various communication systems. In view of the rapid development of communication, there will of course also be future types of communication technology and systems that can embody the present disclosure. The scope of the present disclosure should not be seen as limited to the aforementioned systems only.

[0034] As used herein, the term “network device” refers to a node in a communication network via which terminal devices access the network and receive services therefrom. The network device can refer to a base station (BS) or an access point (AP), e.g., a NodeB (NodeB or NB), an evolved NodeB (eNodeB or eNB), an NR NB (also known as gNB), a remote radio unit (RRU), a radio head (RH), a remote radio head (RRH), a relay, an integrated access and backhaul (IAB) node, a low power node (such as a femto, pico), a non-terrestrial network (NTN) or non-terrestrial network device (such as a satellite network device, low earth orbit (LEO) satellite, and geosynchronous earth orbit (GEO) satellite), an aerial network device, etc., depending on the terminology used and technology applied. In some example embodiments, a radio access network (RAN) split architecture includes a centralized unit (CU) and a distributed unit (DU) at an IAB donor node. The IAB node includes a mobile termination (IAB-MT) part, which behaves like a UE to a parent node, while the DU part of the IAB node behaves like a base station to a next-hop IAB node.

[0035] The term “terminal device” refers to any terminal device capable of wireless communication. As examples, without limitation, a terminal device can also be referred to as a communication device, user equipment (UE), a subscriber station (SS), a portable subscriber station, a mobile station (MS), or an access terminal (AT). A terminal device can include, without limitation, a mobile phone, a cellular phone, a smart phone, a voice over Internet Protocol (VoIP) phone, a wireless local loop phone, a tablet computer, a wearable terminal device, a personal digital assistant (PDA), a portable computer, a desktop computer, an image capture terminal device such as a digital camera, a gaming terminal device, a music storage and playback appliance, a vehicle-mounted wireless terminal device, a wireless endpoint, a mobile station, a laptop-embedded equipment (LEE), a laptop-mounted equipment (LME), a USB dongle, a smart device, a wireless customer-premises equipment (CPE), an Internet of Things (IoT) device, a watch or other wearable device, a head-mounted display (HMD), a vehicle, a drone, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in industrial and / or

[0036] As used herein, the terms “resource,” “transmission resource,” “resource block,” “physical resource block” (PRB), “uplink resource,” or “downlink resource” can refer to any resource used to perform communication (e.g., between a terminal device and a network device), such as a time-domain resource, a frequency-domain resource, a spatial-domain resource, a code-domain resource, or any other combination of time-, frequency, spatial, and / or code-domain resources that support communication, etc. Hereinafter, unless explicitly stated otherwise, resources in the frequency and time domains will be used as examples of transmission resources to describe some example embodiments of the present disclosure. Note that example embodiments of the present disclosure are equally applicable to other resources in other domains.

[0037] Retrieval Augmentation Generation (RAG) is an AI framework to improve the quality of responses generated by an LLM by augmenting the internal representation of information of the LLM based on external knowledge sources (allowing access to information outside of the training data). It can include two stages. One stage is the retrieval stage. In this stage, various algorithms (e.g., similarity scores computed using cosine of user input and external information) are used to search and retrieve pieces of information relevant to the user’s prompt or question. This external information is appended to the user’s prompt and passed to the LLM. The other stage is the generation stage. In the generation stage, the LLM relies on the augmented prompt and its internal representation of the training data to generate an answer; and the final response is a combination of the retrieved information and the model’s own generation capabilities.

[0038] RAG can ensure that LLM models have access to current and reliable relevant facts, thereby improving accuracy and reducing hallucinations. LLM hallucination is a phenomenon where an LLM generates nonsense or is not faithful to the source content. In addition, RAG can reduce the cost of updating LLM models, such as those that occur during the retraining or fine-tuning process.

[0039] LLMs can use one input and output data modality related to language. LMMs can combine various data modalities, such as text, audio, visual, sensor data, and capture the correlation between different modalities. LLMs can be applicable to any type of data, including web data.

[0040] In addition, small language models (SLMs) are computationally less intensive than LLMs in both training and inference, and they have good performance, especially when trained and used for specific problems. Therefore, in addition to LLMs and LMMs, SLMs also have high relevance in telecommunications applications.

[0041] Currently, in 3rd Generation Partnership Project (3GPP) Release 19, there are discussions on Artificial Intelligence Machine Learning (AIML or AI / ML) management. As part of their objectives, these discussions include discussions related to generative AI. For example, it was studied that AI / ML management and operational capabilities can support different types of AI / ML techniques as needed to support AI / ML in 5G systems (5GS), such as federated learning, reinforcement learning, online and offline training, distributed learning, and generative AI.

[0042] Further, solutions supported by LLM, LMM, or SLM provide answers to a consumer's question or prompt during the inference phase by two methods, method 1 that uses internal knowledge and method 2 that uses external knowledge. For example, method 1 can use internal representation of information built into the model parameters and can be learned during the model pre-training and fine-tuning process. The advantage of this method is fast generation of responses and reduced cost as the answer generation relies only on the model internal information without the need for external sources. However, the disadvantage of this method is increased hallucination risk as the model internal information representation can be outdated because the training data used can not correspond to the current state or the model internal information representation can not capture all the information needed to generate the answer.

[0043] Method 2 can employ a RAG to base the output on model external information sources. Such information sources are not used during training and are not captured in the model parameters. Thus, the external information sources can be used as additional information to the model internal representation. Such information sources can include different databases that reside inside or outside of certain ranges (e.g., carrier premises, trusted domains). The advantage of this method is improved answer quality and reduced risk of hallucination as the model relies on recent and additional information that was not seen during training. However, this method can have disadvantages such as constraints in terms of access rights, privacy issues, increased cost, and latency issues related to retrieving information from external sources.

[0044] Thus, during the inference phase, there is a challenge in deciding whether to use internal knowledge or external knowledge, taking into account the advantages and disadvantages described above. However, during the training phase, the challenge is to determine how to partition the available information and store the information as either model internal knowledge or as external knowledge to be retrieved during inference. The more common parts of the information are built into the ML model (e.g., via pre-training or fine-tuning) as internal knowledge, and the more specific to the domain and regularly updated information will be stored outside of the ML model as external knowledge. One consideration is the cost of knowledge partitioning, as it is well understood that pre-training and fine-tuning of large models is very expensive. As knowledge is always changing and updating, it is not cost effective and practical to have all the knowledge built into the large model as internal knowledge.

[0045] Thus, there can be a need for a method to manage the partitioning of knowledge into internal and external knowledge. Further, there can be a need for a method to access internal and external knowledge.

[0046] Example embodiments of the present disclosure provide a solution for the use of external knowledge. In this solution, a device (referred to as a first device), such as a management service (MnS) consumer, sends information related to the use of external knowledge for a communication network service to another device (referred to as a second device), such as a MnS producer. In the context of the present disclosure, the external knowledge is stored external to a communication network function configured to operate the communication network service. Then, after the first device sends a request for the communication network service to the second device, the second device sends a response for the communication network service to the first device, where the response is based on at least one of internal knowledge of the communication network function or the external knowledge.

[0047] This solution enables a MnS consumer to configure how external knowledge is used by a MnS producer. In this way, external knowledge can be used more efficiently, and the efficiency of the communication network service can be improved.

[0048] Figure 1 An example communication environment 100 in which example embodiments of the present disclosure can be implemented is shown. The communication environment 100 can be part of a communication network, including a first device 110 and a second device 120. In some example embodiments, the first device 110 can operate as a terminal device, such as a UE; while the second device 120 can operate as a network device, such as a gNB.

[0049] It should be understood that Figure 1 The number and types of devices shown in FIG. 1 are for illustrative purposes only and are not meant to suggest any limitations. The communication environment 100 can include any suitable number and types of devices for implementing embodiments of the present disclosure.

[0050] In example embodiments where the first device 110 operates as a terminal device and the second device 120 operates as a network device, the transmission direction from the second device 120 to the first device 110 can be referred to as the downlink (DL), and the transmission direction from the first device 110 to the second device 120 can be referred to as the uplink (UL). In the DL, the second device 120 is the transmitting (TX) device (or transmitter), and the first device 110 is the receiving (RX) device (or receiver). In the UL, the first device 110 is the TX device, and the second device 120 is the RX device. In example embodiments where both the first device 110 and the second device 120 operate as terminal devices, the link between the first device 110 and the second device 120 can be referred to as a sidelink (SL). In the SL, one of the first device 110 and the second device 120 is the TX device, and the other of the first device 110 and the second device 120 is the RX device.

[0051] In the following, for the purpose of illustration, some example embodiments are described in which the first apparatus 110 operates as a terminal device and the second apparatus 120 operates as a network device. However, in some example embodiments, operations described with respect to a terminal device can be implemented at a network device or other device, and operations described with respect to a network device can be implemented at a terminal device or other device.

[0052] Communications in the communication environment 100 can be implemented according to any suitable communication protocol, including but not limited to cellular communication protocols of first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), fifth generation (5G), sixth generation (6G), and / or the like, wireless local area network communication protocols such as Institute of Electrical and Electronics Engineers (IEEE) 802.11, and / or any other protocol that is currently known or developed in the future. Moreover, communications can utilize any suitable wireless communication techniques, including but not limited to code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), frequency division duplexing (FDD), time division duplexing (TDD), multiple-input multiple-output (MIMO), orthogonal frequency division multiple access (OFDMA), discrete Fourier transform spread OFDM (DFT-s-OFDM), and / or any other techniques that are currently known or developed in the future.

[0053] In the environment 100, the first apparatus 110 and the second apparatus 120 can communicate with each other to provide various communication network services. In an example, the second apparatus 120 can operate as a MnS producer, which can be a network device deployed in an access network such as a radio access network (RAN) or a core network such as a 5G core (5GC). The first apparatus 110 can operate as a MnS consumer, which can be a terminal device.

[0054] To support the communication network services, the second apparatus 120 can provide a communication network function configured to operate the communication network services. In some example embodiments, the communication network function can be implemented using a machine learning (ML) model, which can include a language-based model, or a model based on other modalities such as nested logit, image, code, and / or the like. In an example, the ML model can be a LLM, a LMM, and a SLM. The communication network function provided by the second apparatus 120 can include a Retrieve Augmented Generation Management Function (RAG-MF), which can provide a RAG management service. The RAG-MF can use external knowledge for RAG.

[0055] Figure 2 An example RAG process 200 according to some example embodiments of the present disclosure is shown. As Figure 2As shown, for example, the RAG framework includes two stages: a retrieval stage 205 and a generation stage 210. During the retrieval stage 205, information 212 related to the user prompt 214 can be retrieved from external knowledge 216 to enrich the user prompt 214. The external knowledge 216 can originate from a database, the internet, or other knowledge repository that is not part of the internal knowledge of the LLM 218. The user prompt 220 enriched with the retrieved information is then input to the LLM 218. During the generation stage 210, the LLM 218 uses the enriched user prompt and its internal knowledge to generate an output or answer 222. Thus, the output or answer 222 is based on both the external knowledge and the internal knowledge of the LLM 218.

[0056] Some example embodiments provide management capabilities for communication network functions. For example, the division of knowledge into internal and external knowledge can be managed. The use of external knowledge during the model inference stage can be activated or deactivated. Configuration capabilities can be provided to control the extent to which external knowledge will be used, e.g., how much data will be retrieved and used for output generation or how much similarity of relevant information is selected. For example, 80% similarity can result in more information being retrieved than 90% similarity, while 90% similarity can indicate that less, more relevant amount of information is used. Further configuration capabilities can be provided to control how external knowledge is used, including contextual or cost considerations. In addition, the use of external knowledge can be reported.

[0057] In addition, some example embodiments provide mechanisms that allow an MnS producer (as an example of a communication network service) that supports an LLM / LMM / SLM service to expose various information for generating an output in response to a prompt by an MnS consumer. The information can include utilization of internal or external knowledge or different knowledge sources, associated implications (such as cost), etc. Mechanisms can also be provided that allow the MnS producer to report whether internal or external knowledge was used in generating the output.

[0058] In addition, some example embodiments provide mechanisms that allow an MnS producer to enable an authorized consumer to activate or deactivate the use of external knowledge and to enable the authorized consumer to configure how external and internal knowledge can be used in generating an output, including the amount of data to be retrieved. In addition, mechanisms are provided that allow an MnS producer to enable an authorized consumer to configure how available knowledge is divided into internal and external knowledge during the pre-training and fine-tuning process. This division can enable automation of the LLM / LMM / SLM fine-tuning or re-training process when new information is generated or collected.

[0059] Reference will be made to the following Figure 3 Some example implementations are described with respect to FIG. 6.

[0060] Figure 3 A signaling diagram illustrating communication between a first apparatus 110 and a second apparatus 120 for configuring use of external knowledge is shown, in accordance with some example embodiments of the present disclosure.

[0061] As shown, the first apparatus 110, such as a MnS consumer, sends 302 information related to use of external knowledge for a communication network service to the second apparatus 120, such as a MnS producer. The communication network service can include any service related to a communication network. Accordingly, the second apparatus 120 receives 304 the information from the first apparatus 110. The external knowledge is stored external to a communication network function configured to operate the communication network service. In some example embodiments, the communication network function can be implemented using a machine learning model, such as a LLM, LMM, or SLM. Figure 3

[0062] Then, the first apparatus 110 sends 306 a request related to the communication network service to the second apparatus 120. After the second apparatus 120 receives 308 the request, the second apparatus 120 sends 310 a response for the communication network service to the first apparatus 110. Accordingly, the first apparatus 110 receives 312 the response for the communication network service. The response is based on at least one of the internal knowledge and the external knowledge of the communication network function. For example, the first apparatus 110 can send a user prompt or question to the second apparatus 120 as the request for the communication network service. After the second apparatus 120 receives the user prompt or question, based on the information related to use of the external knowledge, the second apparatus 120 can generate an answer or output using only its internal knowledge, or only the external knowledge, or both the internal knowledge and the external knowledge.

[0063] In some example embodiments, the information related to use of the external knowledge sent 302 by the first apparatus 110 to the second apparatus 120 can include a request to activate or deactivate use of the external knowledge. For example, the first apparatus 110 can be enabled to configure the second apparatus 120 to use a RAG-MF (which uses external knowledge) to generate an answer to a user prompt or request by enabling activation or deactivation of the RAG-MF and thus use of the external knowledge.

[0064] ​In some example embodiments, activation or deactivation of the use of external knowledge can be requested at different granularities. In an example, activation or deactivation can be requested for a communication network function. For example, RAG activation can be requested for an ML model used to implement a communication network function. In another example, activation or deactivation can be requested for a session or interaction between a customer of a communication network service and a communication network function. For example, RAG activation can be requested for an ongoing session / interaction between a consumer and a model. In yet another example, activation or deactivation can be requested for individual requests input to a communication network function. For example, RAG activation can be requested for each individual request or prompt given to a model.

[0065] In some example embodiments, the information related to the use of external knowledge can comprise information related to at least one external knowledge resource to be used to retrieve external knowledge. In some example embodiments, the information related to the at least one external knowledge resource can comprise an address, a name, a uniform resource locator (URL), and / or a location of the at least one external knowledge resource. For example, the first apparatus 110 can be enabled to specify a desired external source for augmenting answer generation, including specifying an address, a name, a URL, or a location of the at least one external knowledge resource, such as a particular country.

[0066] In some example embodiments, the information related to the use of external knowledge can comprise information related to an amount of external knowledge to be retrieved for a communication network service. In some example embodiments, the information related to the amount of external knowledge can comprise the amount of external knowledge. In an example, the first apparatus 110 can be enabled to configure the second apparatus 120 with information about a target amount and a maximum amount of data that can be retrieved for generating answers. For example, the first apparatus 110 can set an amount of data to be retrieved. In some example embodiments, the information related to the amount of external knowledge can comprise at least one criterion to control the amount of external knowledge. For example, the first apparatus 110 can set a criterion, such as a similarity score, to control the amount of external knowledge.

[0067] In some example embodiments, the information related to the use of external knowledge can comprise at least one policy for the use of external knowledge. In some example embodiments, the at least one policy for the use of external knowledge can comprise constraints and / or conditions for the activation or deactivation of the use of external knowledge. For example, the first device 110 can be enabled to define for the second device 120 a policy containing constraints or conditions for the activation or deactivation of the use of external knowledge under which the RAG and the use of external knowledge are activated. In some example embodiments, the at least one policy for the use of external knowledge can comprise constraints and / or conditions for the switching between different external knowledge sources. For example, the first device 110 can be enabled to define for the second device 120 a policy containing constraints or conditions when one external knowledge source should be switched with another external knowledge source.

[0068] In some example embodiments, the constraints or conditions can be related to a type of the communication network service and / or a network slice type associated with the communication network service, such as the affected service / slice type. Alternatively or additionally, the constraints or conditions can be related to a location of the first device 110 and / or a type of the first device 110. For example, in embodiments where the first device 110 operates as a UE, the conditions and constraints can be related to a UE location and / or a UE type.

[0069] In some example embodiments, the at least one policy for the use of external knowledge can be given in different granularities. For example, the at least one policy can be associated with a communication network function, such as a ML model used to implement the communication network function. Alternatively or additionally, the at least one policy can be associated with a session or interaction between a customer of the communication network service and the communication network function, such as an ongoing session / interaction between a consumer and a model. Alternatively or additionally, the at least one policy can be associated with an individual request input to the communication network function. For example, the at least one policy can be provided for each individual request / prompt given to the model.

[0070] In some example embodiments, the at least one policy for the use of external knowledge can be based on different aspects. In examples, the at least one policy can be based on at least one performance metric measured for the use of external knowledge. For example, when external knowledge is used, latency, monetary and / or energy costs, and other performance metrics can be estimated or measured by the second device 120 (e.g., the MnS producer). The performance metrics can be provided by the second device 120 to the first device 110. Then, the first device 110 can determine the at least one policy for the use of external knowledge based on the performance metrics.

[0071] Alternatively or additionally, the at least one strategy can be based on a confidence level of the external knowledge. The confidence level can be measured by previous experience reviews and / or ratings of other consumers using the source of external knowledge. Alternatively or additionally, the at least one strategy can be based on an availability level, a reliability level, and / or a dependency level of the external knowledge. Alternatively or additionally, the at least one strategy can be based on a task of the communication network service, such as a context of a task addressed by the MnS consumer. Alternatively or additionally, the at least one strategy can be based on a type of a request input to the communication network function, such as a prompt type, e.g., a high-confidentiality prompt or an open-ended prompt.

[0072] In some example embodiments, the second apparatus 120 can expose information related to a capability of the communication network function to support use of external knowledge. For example, the second apparatus 120 can be enabled to expose information about a capability to support the RAG-MF to use external knowledge to provide an output or an answer. Accordingly, the first apparatus 110 can obtain the information from the second apparatus 120. Then, based on the obtained information, the first apparatus 110 can determine information related to use of external knowledge to transmit to the second apparatus 120.

[0073] In some example embodiments, the information related to the capability of the communication network function can comprise an indication that the communication network function does not support use of external knowledge. For example, if the second apparatus 120 does not support the RAG-MF to use external knowledge, but relies on internal knowledge only, the second apparatus 120 can send an indication with a flag “RAG-MF support” set to false.

[0074] In some example embodiments, the information related to the capability of the communication network function can comprise an indication that the communication network function supports use of external knowledge. For example, if the second apparatus 120 supports the RAG-MF and relies on external knowledge, the second apparatus 120 can send an indication with a flag “RAG-MF support” set to true.

[0075] In some example embodiments, the information related to the capabilities of the communication network function can comprise a form of external knowledge. For example, the information can comprise a description of the external knowledge, which can indicate a type of the external knowledge, e.g. whether the external knowledge originates from a vector database, a knowledge graph, a relational database, a normal database, or a structured data set. Alternatively or additionally, the information related to the capabilities of the communication network function can comprise an application of the external knowledge, such as a domain of the external knowledge, like mobility optimization, coverage optimization, network tracking, etc. Alternatively or additionally, the information related to the capabilities of the communication network function can comprise a storage address or area of the external knowledge, e.g. an address or area where the information is stored. Alternatively or additionally, the information related to the capabilities of the communication network function can comprise a description of a constraint or recommendation for using an external knowledge source of the at least one external knowledge source, e.g. a description of a constraint or recommendation for using certain knowledge sources.

[0076] In some example embodiments, the information related to the capabilities of the communication network function can comprise at least one entity owning the external knowledge. For example, the entity owning the external knowledge can be the same operator or a third party.

[0077] In some example embodiments, the information related to the capabilities of the communication network function can comprise at least one performance metric measured for the usage of the external knowledge. For example, the performance metric can comprise an energy and / or monetary cost, e.g. due to a charging policy for invoking an external application programming interface (API) or a delay caused by the usage of the external knowledge.

[0078] In some example embodiments, the second device 120 can send a report on the usage of the external knowledge to the first device 110. Then, the first device 110 can receive the report on the usage of the external knowledge. The report can comprise any information related to the usage of the external knowledge by the second device 120.

[0079] In some example embodiments, the report on the usage of the external knowledge can comprise an indication of whether the external knowledge was used for the communication network service. For example, the second device 120 can be enabled to report whether the RAG-MF and the external knowledge were used when generating a response to a request or a hint. Alternatively or additionally, the report on the usage of the external knowledge can comprise at least one performance metric (such as a delay and / or a cost) measured for the usage of the external knowledge, e.g. which delay or energy or monetary cost has been encountered when the external knowledge was used.

[0080] Alternatively or additionally, the report on usage of external knowledge can include at least one external knowledge source that has been used for the communication network service. For example, the first device 110, which can operate as an MnS consumer, can receive only information on past usage of external knowledge sources as part of the usage report. Alternatively or additionally, the report on usage of external knowledge can include at least one external knowledge source to be used for the communication network service. For example, the second device 120, which can operate as an MnS producer, can have the ability to discover new external knowledge sources even in a session between a customer and a model, and also report usage of these external sources. In some example embodiments, the first device 110 can indicate, as part of its configuration for external knowledge, that a part of a session can be addressed based on a new external knowledge source if the second device 120 has discovered the new external knowledge source that complies with the cost, latency, etc. limits. The remaining part of the session can need to be addressed by the already used or discovered external knowledge source.

[0081] In some example embodiments, the report on usage of external knowledge can be provided at different granularities. For example, the report can be associated with a communication network function, such as an ML model used to implement the communication network function. Alternatively or additionally, the report can be associated with a session or interaction between a customer of a communication network service and a communication network function, such as an ongoing session / interaction between a customer and a model. Alternatively or additionally, an individual request is input to the communication network function. For example, a report can be provided for each individual request / prompt given to the model.

[0082] In some example embodiments, the first device 110 can send to the second device 120 a preference for a split of internal and external knowledge for the communication network function. The preference for the split can be based on the report on usage of external knowledge received from the second device 120. For example, to implement a pre-training or fine-tuning of a model, the first device 110 can indicate a preference for the split of knowledge. That is, the first device 110 can indicate which part of the knowledge can be built in the model parameters as internal knowledge, and which part of the knowledge can be stored outside the model as external knowledge.

[0083] In some example embodiments, the preference for the split can include characteristics of the internal knowledge to be built in the communication network function. For example, during a pre-training or fine-tuning process, the first device 110 can specify characteristics of the internal knowledge to be obtained or characteristics of the data to be used to build the internal knowledge, e.g. data type, data volume, data range such as geographical area, network functions from which data should be collected, time period for which data should be collected, etc. The specified characteristics can be given as a parameter named “build-in knowledge” with different values such as “public” or “public and private”.

[0084] In some example embodiments, the first device 110 can send to the second device 120 preferences for further partitioning of the internal knowledge and / or further partitioning of the external knowledge. Thus, the first device 110 (e.g., as a MnS consumer) can experience only the prompt enhancement based on a smaller internal / external knowledge base. A trade-off between performance and cost can be achieved.

[0085] In some example embodiments, the second device 120 can generate a recommendation for a partitioning of the internal knowledge and the external knowledge for the communication network function. The communication network function (e.g., the ML model) can be pre-trained or fine-tuned based on the recommendation. The recommendation can be generated based on an analysis of a relationship between usage of the external knowledge and at least one performance metric of the communication network service. The at least one performance metric of the communication network service can include statistical and / or predicted performance metrics of the communication network service in case the external knowledge is used. In some example embodiments, the second device 120 can send such analysis to the first device 110. The analysis can also be included in a report on usage of the external knowledge.

[0086] For example, the second device 120 can collect reports on RAG-MF and external knowledge usage and correlate them with other performance metrics of the ML model, such as quality of response, delay in generating a response, and energy or monetary cost. Based on such information, the second device 120 can provide statistics and predictions of the relationship between RAG-MF usage and answer quality, delay, and cost. Based on the statistics and predictions, the second device 120 can generate a recommendation for a partitioning of the internal knowledge and the external knowledge. The statistics and predictions can also be disclosed to the first device 110 as additional information to configure the RAG-MF.

[0087] In some example embodiments, the second device 120 can register the communication network function to a third device (which can be a ML model repository) with information related to the capability of the communication network function to support usage of the external knowledge. For example, the second device 120 can register the ML model with the capability to support RAG-MF and details on the retrieval process, such as available knowledge sources and available information on RAG analysis. In some example embodiments, the first device 110 can obtain the information related to the capability of the communication network function to support usage of the external knowledge from the third device.

[0088] Example procedures for RAG configuration and reporting will be described below with reference to Figure 4 An example procedure for RAG configuration and reporting will be described below with reference to

[0089] As Figure 4As shown, at 401, the MnS producer 420 can expose the capability to support RAG-MF and use of external knowledge to request or hint the generation of output or answers to the consumer. For example, if the MnS producer 420 does not support RAG-MF but relies only on internal knowledge captured in the model parameters, the MnS producer 420 can send an indication to the MnS consumer 410 that the use of RAG-MF and external knowledge is not supported by the MnS producer 420. The indication can be a dedicated flag named “RAG support” that is marked as false. In another example, if the MnS producer 420 supports RAG and relies on external knowledge, the MnS producer 420 can send an indication to the MnS consumer 410 that the use of RAG-MF and external knowledge is supported by the MnS producer 420. The indication can be a dedicated flag named “RAG support” that is marked as true.

[0090] Alternatively or additionally, if the MnS producer 420 supports RAG-MF, the MnS producer 420 can expose further information related to the capability of RAG at 401. This can include: a description of the external knowledge that can be retrieved during the RAG process, e.g., the type of knowledge, such as a knowledge graph, a vector database, a normal database; information related to the knowledge sources that can be used during the RAG process, such as the address or area where the information is stored, the entity that owns the knowledge source; information about the related costs in terms of energy consumption, monetary expenditure, or latency when using RAG-MF; information about constraints or recommendations for using certain knowledge sources, e.g., to which contexts (service or UE location or type) the constraints or recommendations can apply.

[0091] At 403, after the MnS consumer 410 receives the information related to the capability to support RAG-MF and use of external knowledge, the MnS consumer 410 can configure the MnS producer so that RAG-MF and external knowledge are used in the desired way when generating answers. For example, if it is desired to use RAG-MF when generating answers, the MnS consumer 410 can configure the flag “RAG activation” to true. Alternatively, the MnS consumer 410 can request RAG-MF to be activated and specify further details, such as information about the knowledge sources to be used, such as the address, the name of the entity, or the target and maximum amount of data to be retrieved. RAG activation can be requested at different granularities, including the ML model, the session or interaction between the customer of the communication network service and the ML model, or individual requests to the ML model.

[0092] At 405, alternatively or in addition to step 403, the MnS consumer can provide a policy that defines under which conditions the RAG-MF can be activated or deactivated, or when there is a switch between different knowledge sources. For example, depending on the expected cost in terms of energy consumption, monetary expenditure, or latency when generating an answer, the RAG-MF can be activated or deactivated, or there can be a switch between different knowledge sources (e.g., a closer source belonging to the same operator or a cheaper source from a third-party provider, etc.). The policy can be provided at different granularities, including the ML model, a session or interaction between a customer of the communication network service and the ML model, or an individual request that is input to the ML model.

[0093] At 407, the MnS producer 420 can report the RAG-MF usage for generating answers, including whether the RAG-MF was used, which knowledge sources were applied, which latencies or energy or monetary costs have been encountered. Alternatively or additionally, the MnS consumer 410 can read this information from the ML model management object. The reporting can be provided at different granularities, including the ML model, a session or interaction between a customer of the communication network service and the ML model, or an individual request that is input to the ML model.

[0094] At 409, the MnS producer 420 can collect all information about RAG usage and correlate it with other ML model related metrics, such as latency and energy or monetary costs when generating responses, and provide an analysis or statistics or predictions about the relationship between RAG-MF usage and model performance (such as costs) when generating responses.

[0095] At 411, the information derived in step 405 can be used to enhance the information reported or exposed to the MnS consumer 410, so that the consumer can best configure the RAG-MF usage.

[0096] At 413, the MnS producer 420 can register the ML model and its capabilities to support RAG and details about the retrieval process, such as available knowledge sources and available information about RAG-MF analysis. Furthermore, based on the RAG-MF analysis done in step 409, the most useful external knowledge sources or RAGs tailored for the usage of the LLM can be identified and stored in the ML model repository 430 as metadata for a given LLM / LMM / SLM version.

[0097] Next, some use cases for a communication network service will be described below with reference to Figures 5A to 6B and Figure 5A and Figure 5B Example use cases regarding intent-based management with and without RAG activation are shown according to some example embodiments of the present disclosure. As Figure 5AIn the illustrated scenario 500A, UE 502 (as an example of the first apparatus 110) can provide a network intent to a model 504 (such as a LLM, SLM, or LMM) to improve network reliability, for example, by minimizing downtime of network functions (NFs) that are components of the network infrastructure. UE 502 can be a service-consuming MnS and can activate or deactivate RAG support. Model 504 can provide recommendations on actions that need to be taken in order to prevent failures and recover from occurred failures as output.

[0098] As Figure 5A illustrated, UE 502 activates RAG support, and thus model 504 can utilize an external knowledge base that can contain up-to-date information, such as customer documents about the NFs deployed in the network and knowledge about ticketed solutions from customer support teams. This knowledge base can be used as external knowledge during the RAG process. The retrieved external knowledge can include troubleshooting tickets and customer documents that are enriched with up-to-date information about the NFs and ticketed solutions. This enriched knowledge is then used to inform the recommendations. Based on the enriched knowledge and analysis performed by model 504 (such as a LLM / SLM / LMM), model 504 outputs a set of actions. These actions can be specifically tailored to minimize NF failures and provide guidance on how to recover from any failures that have already occurred.

[0099] Further, UE 502 can evaluate the recommendations and decide on the actions. UE 502 can also provide feedback to model 504, which can be used to further refine the recommendations and improve the performance of model 504 over time. Model 504 can be designed to learn from updates in customer documents and recent ticketed solutions, ensuring that the recommendations it provides are as reliable and up-to-date as possible.

[0100] Conversely, in scenario 500B as Figure 5B illustrated, UE 502 deactivates RAG support, and thus model 504 can generate a set of actions based only on the network intent and the model’s internal knowledge.

[0101] Figure 6A and Figure 6B Example use cases of performing UE-specific tasks with and without RAG activation are shown in accordance with some example embodiments of the present disclosure. In scenario 500A as Figure 6AIn the illustrated scenario 600A, the UE 502 can initiate the process by submitting a prompt to the model 504, such as a LLM or SLM. In this case, the prompt is a request for a voice-guided tour of the best route for sightseeing in the city. In the case of a LLM as the model 504, the UE 504 can send the prompt to the LLM deployed in the cloud. In the case of an SLM as the model 504, the model 504 can be directly deployed at the UE 502, and the UE 502 can interact with the model 504 on the device. In both cases, the UE 502 is the MnS consumer of the service.

[0102] Depending on the level of satisfaction with the provided answer, the UE 502 can decide whether to activate or deactivate the RAG. For example, to minimize cost, the UE 502 can first choose to deactivate the RAG. If the suggested route for sightseeing is not up-to-date with the latest landmarks or events in the city, the UE 502 can activate the RAG and obtain the latest output.

[0103] As Figure 6A illustrated, the UE 502 activates the RAG support, so the model 504, such as a LLM or SLM, can access external knowledge sources that are updated in real-time. For example, the external knowledge sources can include updated web pages, new internet landscapes for a guide to city A, a travel guide to city B, updates during a sporting event, updates to traffic schedules, and road closures due to sporting events. The model 504, such as a LLM or SLM, can retrieve information related to the user request from these external knowledge sources. The model 504 can utilize the latest information to enrich the output, such as avoiding road closures and congested areas due to events like sporting events. The UE 502 can evaluate the output and decide whether the information is satisfactory. If the provided route does not currently have the latest landmarks or events, the UE 502 can activate the RAG to receive a more accurate and updated guide.

[0104] In the scenario 600B as Figure 6B illustrated, the UE 502 deactivates the RAG support, which means the model 504 can not use external knowledge to generate the response. This can be a decision made by the UE 502 to minimize cost or due to other considerations. The model 504 can provide an output that is a suggested sightseeing route for city A. However, since the RAG is deactivated, the route can not take into account recent changes, such as those due to sporting events.

[0105] In another example use case, the ML MnS consumer can be, for example, a UE hosting an SLM, and the ML MnS producer can be on the network operations, maintenance and management (OAM) side. The purpose of consuming the ML MnS service can be for the UE to be able to access a trusted RAG containing relevant domain-specific data (e.g., related to vehicle UE navigation or local landmarks) when the UE is traversing a particular geographic area. The ML MnS producer can then assist in discovering such RAGs and expose them to the UE. In other words, this RAG discovery process can also be triggered each time the UE is tailored to another slice. Thus, the SLM can be adapted to network slice or application needs by being augmented with appropriate external data by the RAG for the respective network slice. From a RAG instance deployment perspective, these network slices or domain-specific RAGs can be deployed at the edge server or in the cloud.

[0106] An example implementation of RAG configuration and reporting is discussed below with 3GPP SA5 as an example.

[0107] Figure 7 A flowchart of an example method 700 implemented at a first device is shown in accordance with some example embodiments of the present disclosure. For purposes of discussion, the method 700 will be described from the perspective of the first device 110 in Figure 1

[0108] At block 710, the first device 110 sends, to a second device, information related to use of external knowledge for a communication network service. The external knowledge is stored external to a communication network function, and the communication network function is configured to operate the communication network service.

[0109] At block 720, the first device 110 sends, to the second device, a request for the communication network service.

[0110] At block 730, in response to sending the request, the first device 110 receives, from the second device, a response for the communication network service. The response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0111] In some example embodiments, the first device 110 can obtain, from the second device or a third device, information related to a capability of the communication network function to support use of the external knowledge.

[0112] In some example embodiments, the information related to the capability of the communication network function can include: an indication that use of the external knowledge is not supported by the communication network function; or an indication that use of the external knowledge is supported by the communication network function.

[0113] ​In some example embodiments, the information related to the capability of the communication network function can comprise at least one of: a form of the external knowledge, an application of the external knowledge, a storage address or area of the external knowledge, at least one entity that owns the external knowledge, a description of a constraint or recommendation for use of an external knowledge source of the at least one external knowledge source, at least one performance metric of a measure of use of the external knowledge.

[0114] In some example embodiments, the information related to the use of the external knowledge can comprise at least one of: a request for activation or deactivation of the use of the external knowledge; information related to at least one external knowledge source to be used to retrieve the external knowledge; information related to an amount of the external knowledge to be retrieved for the communication network service; and at least one policy for the use of the external knowledge.

[0115] In some example embodiments, the activation or deactivation of the use of the external knowledge is requested for at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0116] In some example embodiments, the information related to the at least one external knowledge source can comprise at least one of an address, a name, a uniform resource locator, and a location of the at least one external knowledge source.

[0117] In some example embodiments, the information related to the amount of the external knowledge comprises at least one of: the amount of the external knowledge, and at least one criterion to control the amount of the external knowledge.

[0118] In some example embodiments, the at least one policy for the use of the external knowledge can comprise at least one of a constraint and a condition for at least one of: the activation or deactivation of the use of the external knowledge; and a switching between different external knowledge sources.

[0119] In some example embodiments, the at least one of the constraint and the condition can be related to at least one of: a type of the communication network service, a network slice type associated with the communication network service, a location of the first apparatus, and a type of the first apparatus.

[0120] In some example embodiments, the at least one policy for the use of the external knowledge is associated with at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0121] In some example embodiments, the at least one policy for the use of external knowledge can be based on at least one of: at least one performance metric of a measure of the use of external knowledge; at least one of a confidence level, an availability level, a reliability level, and a dependency level of the external knowledge; a task of the communication network service; and a type of request input to the communication network function.

[0122] In some example embodiments, the first apparatus 110 can receive, from the second apparatus, a report on the use of external knowledge.

[0123] In some example embodiments, the report on the use of external knowledge comprises at least one of: an indication of whether external knowledge was used for the communication network service; at least one source of external knowledge that has been used for the communication network service; at least one source of external knowledge to be used for the communication network service; and at least one performance metric of a measure of the use of external knowledge.

[0124] In some example embodiments, the report on the use of external knowledge can be associated with at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0125] In some example embodiments, the first apparatus can send, to the second apparatus, a preference for a partitioning of internal knowledge and external knowledge for the communication network function; wherein the preference for the partitioning is based on the received report on the use of external knowledge.

[0126] In some example embodiments, the preference for the partitioning can comprise a characteristic of internal knowledge to be built in the communication network function.

[0127] In some example embodiments, the first apparatus can receive, from the second apparatus, an analysis of a relationship between the use of external knowledge and at least one performance metric of the communication network service.

[0128] In some example embodiments, the communication network function can be implemented using a machine learning model.

[0129] In some example embodiments, the machine learning model comprises a language-based model.

[0130] Figure 8 A flowchart of an example method 800 implemented at the second apparatus is shown in accordance with some example embodiments of the present disclosure. For purposes of discussion, the method 800 will be described from the perspective of the second apparatus 120 in Figure 1

[0131] ​At block 810, the second apparatus 120 receives, from the first apparatus, information related to use of external knowledge for a communication network service. The external knowledge is stored external to the communication network function, and the communication network function is configured to operate the communication network service.

[0132] At block 820, the second apparatus 120 receives, from the first apparatus, a request for the communication network service.

[0133] At block 830, in response to receiving the request, the second apparatus 120 sends, to the first apparatus, a response for the communication network service. The response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0134] In some example embodiments, the second apparatus 120 can expose information related to a capability of the communication network function to support use of external knowledge.

[0135] In some example embodiments, the information related to the capability of the communication network function to support use of external knowledge can include: an indication that use of external knowledge is not supported by the communication network function; or an indication that use of external knowledge is supported by the network function.

[0136] In some example embodiments, the information related to the capability of the communication network function to support use of external knowledge can include at least one of: a form of the external knowledge, an application of the external knowledge, a storage address or area of the external knowledge, at least one entity that owns the external knowledge, a description of a constraint or recommendation to use an external knowledge source of the at least one external knowledge source, at least one performance metric of a measure for use of the external knowledge.

[0137] In some example embodiments, the information related to use of external knowledge can include at least one of: a request for activation or deactivation of use of the external knowledge; information related to at least one external knowledge source to be used to retrieve the external knowledge; information related to an amount of the external knowledge to be retrieved for the communication network service; and at least one policy for use of the external knowledge.

[0138] In some example embodiments, the activation or deactivation of use of the external knowledge can be requested for at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0139] In some example embodiments, the information related to the at least one external knowledge source can include at least one of an address, a name, a uniform resource locator, and a location of the at least one external knowledge source.

[0140] In some example embodiments, the information related to the amount of external knowledge can comprise at least one of: the amount of external knowledge, and at least one criterion to control the amount of external knowledge.

[0141] In some example embodiments, the at least one policy for the use of external knowledge comprises at least one of a constraint and a condition for at least one of: activation or deactivation of the use of external knowledge; and switching between different sources of external knowledge.

[0142] In some example embodiments, the at least one of a constraint and a condition is related to at least one of: a type of the communication network service, a network slice type associated with the communication network service, a location of the first device, and a type of the first device.

[0143] In some example embodiments, the at least one policy for the use of external knowledge can be associated with at least one of: a communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0144] In some example embodiments, the at least one policy for the use of external knowledge is based on at least one of: at least one performance metric measured for the use of external knowledge; at least one of a confidence level, an availability level, a reliability level, and a dependency level of external knowledge; a task of the communication network service; and a type of request input to the communication network function.

[0145] In some example embodiments, the second device 120 can send, to the first device, a report on the use of external knowledge.

[0146] In some example embodiments, the report on the use of external knowledge can comprise at least one of: an indication of whether external knowledge was used for the communication network service; at least one source of external knowledge that has been used for the communication network service; at least one source of external knowledge to be used for the communication network service; and at least one performance metric measured for the use of external knowledge.

[0147] In some example embodiments, the report on the use of external knowledge can be associated with at least one of: a communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0148] In some example embodiments, the second device 120 can receive, from the first device, a preference for a split of internal knowledge and external knowledge for the communication network function after sending the report on the use of external knowledge.

[0149] In some example embodiments, preferences for partitioning may include characteristics of the internal knowledge to be built into the communication network functionality.

[0150] In some example embodiments, the second device 120 can generate recommendations for the division of internal and external knowledge for communication network functions; and pre-train or fine-tune communication network functions based on the recommendations.

[0151] In some example embodiments, the second device 120 may send an analysis to the first device regarding the relationship between the use of external knowledge and at least one performance metric of the communication network service.

[0152] In some example embodiments, the second device 120 may register communication network functions and related information about the ability of the communication network functions to support the use of external knowledge with the third device.

[0153] In some example implementations, machine learning models can be used to implement communication network functions.

[0154] In some example embodiments, a first means capable of performing any of the methods in method 700 (e.g., Figure 1 The first device 110 may include components for performing the corresponding operations of method 700. These components may be implemented in any suitable form. For example, the components may be implemented in a circuit or software module. The first device may be implemented as or included in... Figure 1 In the first device 110.

[0155] In some example embodiments, the first device includes: a component for sending information to a second device relating to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside the communication network function and the communication network function is configured to operate the communication network service; a component for sending a request for the communication network service to the second device; and a component for receiving a response for the communication network service from the second device in response to sending the request, wherein the response is based on at least one of: internal knowledge of the communication network function and external knowledge.

[0156] In some example embodiments, the first device may further include a component for obtaining information from the second or third device related to the ability to use external knowledge that supports the communication network functions.

[0157] In some example embodiments, information related to the capabilities of the communication network function includes: an indication that the use of external knowledge is not supported by the communication network function; or an indication that the use of external knowledge is supported by the communication network function.

[0158] In some example embodiments, the information related to the capability of the communication network function comprises at least one of: a form of the external knowledge, an application of the external knowledge, a storage address or area of the external knowledge, at least one entity that owns the external knowledge, a description of a constraint or recommendation for use of an external knowledge source of the at least one external knowledge source, at least one performance metric measured for use of the external knowledge.

[0159] In some example embodiments, the information related to the use of the external knowledge comprises at least one of: a request for activation or deactivation of the use of the external knowledge; information related to at least one external knowledge source to be used to retrieve the external knowledge; information related to an amount of the external knowledge to be retrieved for the communication network service; and at least one policy for the use of the external knowledge.

[0160] In some example embodiments, the activation or deactivation of the use of the external knowledge is requested for at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0161] In some example embodiments, the information related to the at least one external knowledge source can comprise at least one of an address, a name, a uniform resource locator, and a location of the at least one external knowledge source.

[0162] In some example embodiments, the information related to the amount of the external knowledge comprises at least one of: the amount of the external knowledge, and at least one criterion to control the amount of the external knowledge.

[0163] In some example embodiments, the at least one policy for the use of the external knowledge can comprise at least one of a constraint and a condition for at least one of: the activation or deactivation of the use of the external knowledge; and a switching between different external knowledge sources.

[0164] In some example embodiments, the at least one of the constraint and the condition can be related to at least one of: a type of the communication network service, a network slice type associated with the communication network service, a location of the first apparatus, and a type of the first apparatus.

[0165] In some example embodiments, the at least one policy for the use of the external knowledge is associated with at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0166] In some example embodiments, the at least one policy for the use of external knowledge can be based on at least one of: at least one performance metric of a measure of the use of external knowledge; at least one of a confidence level, an availability level, a reliability level, and a dependency level of the external knowledge; a task of the communication network service; and a type of request input to the communication network function.

[0167] In some example embodiments, the first apparatus further comprises means for receiving, from the second apparatus, a report on the use of external knowledge.

[0168] In some example embodiments, the report on the use of external knowledge comprises at least one of: an indication of whether external knowledge was used for the communication network service; at least one source of external knowledge that has been used for the communication network service; at least one source of external knowledge to be used for the communication network service; and at least one performance metric of a measure of the use of external knowledge.

[0169] In some example embodiments, the report on the use of external knowledge can be associated with at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0170] In some example embodiments, the first apparatus further comprises means for sending, to the second apparatus, a preference for a partitioning of internal knowledge and external knowledge for the communication network function, wherein the preference for the partitioning is based on the received report on the use of external knowledge.

[0171] In some example embodiments, the preference for the partitioning can comprise a characteristic of internal knowledge to be built in the communication network function.

[0172] In some example embodiments, the first apparatus further comprises means for receiving, from the second apparatus, an analysis of a relationship between the use of external knowledge and at least one performance metric of the communication network service.

[0173] In some example embodiments, the communication network function can be implemented using a machine learning model.

[0174] In some example embodiments, the machine learning model comprises a language-based model.

[0175] In some example embodiments, a second apparatus (e.g., the second apparatus 120 in Figure 1 may comprise means for performing the respective operations of the method 800. The means can be implemented in any suitable form. For example, they can be implemented in circuitry or software modules. The second apparatus can be implemented as or included in the second apparatus 120 in Figure 1 .

[0176] In some example embodiments, the second apparatus comprises: means for receiving, from the first apparatus, information related to the use of external knowledge for a communication network service, wherein the external knowledge is stored external to a communication network function and the communication network function is configured to operate the communication network service; means for receiving, from the first apparatus, a request for the communication network service; and means for sending, to the first apparatus, a response for the communication network service in response to receiving the request, wherein the response is based on at least one of: internal knowledge of the communication network function, and the external knowledge.

[0177] In some example embodiments, the second apparatus further comprises: means for exposing information related to a capability of the communication network function to support the use of external knowledge.

[0178] In some example embodiments, the information related to the capability of the communication network function to support the use of external knowledge can comprise: an indication that the use of external knowledge is not supported by the communication network function; or an indication that the use of external knowledge is supported by the network function.

[0179] In some example embodiments, the information related to the capability of the communication network function to support the use of external knowledge can comprise at least one of: a form of the external knowledge, an application of the external knowledge, a storage address or area of the external knowledge, at least one entity that owns the external knowledge, a description of a constraint or recommendation to use an external knowledge source of the at least one external knowledge source, at least one performance metric of a measure for the use of the external knowledge.

[0180] In some example embodiments, the information related to the use of external knowledge can comprise at least one of: a request for activation or deactivation of the use of the external knowledge; information related to at least one external knowledge source to be used to retrieve the external knowledge; information related to an amount of the external knowledge to be retrieved for the communication network service; and at least one policy for the use of the external knowledge.

[0181] In some example embodiments, the activation or deactivation of the use of the external knowledge can be requested for at least one of: the communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0182] In some example embodiments, the information related to the at least one external knowledge source can comprise at least one of: an address, a name, a uniform resource locator, and a location of the at least one external knowledge source.

[0183] In some example embodiments, the information related to the amount of the external knowledge can comprise at least one of: the amount of the external knowledge, and at least one criterion to control the amount of the external knowledge.

[0184] In some example embodiments, the at least one policy for the use of external knowledge comprises at least one of a constraint and a condition for at least one of: activation or deactivation of the use of external knowledge; and switching between different sources of external knowledge.

[0185] In some example embodiments, the at least one of a constraint and a condition relates to at least one of: a type of the communication network service, a network slice type associated with the communication network service, a location of the first device, and a type of the first device.

[0186] In some example embodiments, the at least one policy for the use of external knowledge can be associated with at least one of: a communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0187] In some example embodiments, the at least one policy for the use of external knowledge is based on at least one of: at least one performance metric measured for the use of external knowledge; at least one of a confidence level, an availability level, a reliability level, and a dependency level of external knowledge; a task of the communication network service; and a type of a request input to the communication network function.

[0188] In some example embodiments, the second device further comprises means for sending, to the first device, a report on the use of external knowledge.

[0189] In some example embodiments, the report on the use of external knowledge can comprise at least one of: an indication of whether external knowledge was used for the communication network service; at least one source of external knowledge that has been used for the communication network service; at least one source of external knowledge to be used for the communication network service; and at least one performance metric measured for the use of external knowledge.

[0190] In some example embodiments, the report on the use of external knowledge can be associated with at least one of: a communication network function, a session or interaction between a customer of the communication network service and the communication network function, and an individual request input to the communication network function.

[0191] In some example embodiments, the second device further comprises means for receiving, from the first device, a preference for a partitioning of internal knowledge and external knowledge for the communication network function after sending the report on the use of external knowledge.

[0192] In some example embodiments, the preference for the partitioning can comprise a characteristic of internal knowledge to be built in the communication network function.

[0193] In some example embodiments, the second apparatus further comprises means for generating a recommendation of a partitioning of internal knowledge and external knowledge for the communication network function; and means for pre-training or fine-tuning the communication network function based on the recommendation.

[0194] In some example embodiments, the second apparatus further comprises means for sending, to the first apparatus, an analysis of a relationship between the use of the external knowledge and at least one performance metric of the communication network service.

[0195] In some example embodiments, the second apparatus further comprises means for registering, with a third apparatus, the communication network function and information related to a capability of the communication network function to support the use of the external knowledge.

[0196] In some example embodiments, the communication network function can be implemented using a machine learning model.

[0197] Figure 9 is a simplified block diagram of a device 900 suitable for implementing example embodiments of the present disclosure. The device 900 can be provided to implement a communication device, for example, the first apparatus 110 or the second apparatus 120 as shown in Figure 1 FIG. 1. As shown, the device 900 includes one or more processors 910, one or more memories 920 coupled to the processors 910, and one or more communication modules 940 coupled to the processors 910.

[0198] The communication module 940 is for bidirectional communication. The communication module 940 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interface can represent any interface necessary to communicate with other network elements. In some example embodiments, the communication module 940 can include at least one antenna.

[0199] The processor 910 can be of any type suitable to the local technical network, and can include one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multi-core processor architectures, as non-limiting examples. The device 900 can have multiple processors such as a special-purpose integrated circuit chip that is time-slaved to a clock that is synchronized with a master processor.

[0200] The memory 920 can include one or more non-transitory memories and one or more volatile memories. Examples of non-transitory memories include, but are not limited to, read only memories (ROMs) 924, electrically programmable read only memories (EPROMs), flash memories, hard disks, compact disks (CDs), digital video disks (DVDs), optical disks, laser disks, and other magnetic and / or optical storage. Examples of volatile memories include, but are not limited to, random access memories (RAMs) 922 and other volatile memories that do not persist during power down.

[0201] The computer program 930 comprises computer-executable instructions for performing the operations / actions of some example embodiments of the present disclosure, which are executed by the associated processor 910. The instructions of the program 930 can include instructions for performing the operations / actions of some example embodiments of the present disclosure. The program 930 can be stored in a memory, such as the ROM 924. The processor 910 can perform any suitable actions and processes by loading a program 930 into the RAM 922.

[0202] Example embodiments of the present disclosure can be implemented with the aid of the program 930, such that the apparatus 900 can perform any process of the present disclosure as discussed with reference to Figures 1 to 8 Example embodiments of the present disclosure can also be implemented by hardware or by a combination of software and hardware.

[0203] In some example embodiments, the program 930 can be tangibly embodied in a computer-readable medium, which can include a memory in the apparatus 900, such as the memory 920, or other storage devices accessible by the apparatus 900. The apparatus 900 can load the program 930 from the computer-readable medium into the RAM 922 for execution. In some example embodiments, the computer-readable medium can include any type of non- transitory storage medium, such as a ROM, an EPROM, flash memory, a hard disk, a CD-ROM, a DVD, and the like. The term “non-transitory” as used herein is a limitation of the medium itself (i.e., tangible, rather than a signal), and not a limitation of the durability of the data storage (e.g., RAM vs. ROM).

[0204] Figure 10 An example of a computer-readable medium 1000, which can be in the form of a CD, DVD, or other optical storage disk, is shown. The computer-readable medium 1000 has the program 930 stored thereon.

[0205] In general, the various embodiments of the present disclosure can be implemented in hardware or special-purpose circuits, software, logic or any combination thereof. Some aspects can be implemented in hardware, and other aspects can be implemented in firmware or software which can be executed by a controller, microprocessor or other computing device. While various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein can be implemented in hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controler or other computing devices, or some combination thereof.

[0206] Some example embodiments of the present disclosure also provide at least one computer program product tangibly stored on a computer readable medium, such as a non-transitory computer readable medium. The computer program product includes computer executable instructions, such as those included in program modules, executed by devices in a target physical or virtual processor to perform any of the methods as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The functionality of the program modules can be combined or split between program modules as desired in various embodiments. Machine executable instructions for program modules can be executed within a local or distributed device. In a distributed device, program modules can be located in both local and remote storage media.

[0207] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, causes the machine to perform the functions / operations specified in the flow diagrams and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, as a stand-alone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0208] In the context of the present disclosure, computer program code or related data can be embodied by any suitable carrier wave, including a signal, computer readable medium, etc.

[0209] The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0210] Moreover, although operations have been depicted in a particular order, this should not be understood as requiring such order nor that all illustrated operations be performed to achieve desirable results. In certain circumstances, multitasking and parallel processing can be advantageous. Likewise, although several specific implementation details have been included for purpose of disclosure, these should not be construed as limitations of the scope of the disclosure, but rather as descriptions of features that can be specific to particular embodiments. Certain features that are described in the context of separate embodiments can also be implemented in combination with each other. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments, separately or in any suitable sub-combination.

[0211] Although the present disclosure has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject of the appended claims defines the scope of protection of the disclosure and is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as example forms of implementing the claims.

Claims

1. A first means for communication, the first means comprising: At least one processor; as well as At least one memory storing instructions that, when executed by the at least one processor, cause the first device to at least: Send information to a second device relating to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside of a communication network function and the communication network function is configured to operate the communication network service; Send a request for the communication network service to the second device; as well as In response to sending the request, a response for the communication network service is received from the second device. The response is based on at least one of the following: internal knowledge of the communication network functions, and external knowledge.

2. The first apparatus according to claim 1, wherein the at least one memory and the at least one processor further enable the first apparatus to: Information related to the ability of the communication network function to support the use of the external knowledge is obtained from the second or third device.

3. The first apparatus of claim 2, wherein the information relating to the capability of the communication network function includes: The indication that the use of the external knowledge is not supported by the communication network function; or The use of the external knowledge is indicated by the communication network functionality.

4. The first apparatus of claim 2, wherein the information relating to the capability of the communication network function includes at least one of the following: The form of the external knowledge. The application of the aforementioned external knowledge The storage address or region of the external knowledge. At least one entity possessing the aforementioned external knowledge. Use at least one external knowledge source's constraints or recommended description. At least one performance metric for the use of the external knowledge.

5. The first device according to any one of claims 1 to 3, wherein the information related to the use of the external knowledge includes at least one of the following: Requests for activation or deactivation of the external knowledge in relation to its use; Information related to at least one external knowledge resource to be used to retrieve the external knowledge; Information relating to the amount of external knowledge to be retrieved for use in the communication network services; as well as At least one strategy used in response to the external knowledge.

6. The first apparatus of claim 5, wherein the activation or deactivation of the use of the external knowledge is requested for at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

7. The first apparatus of claim 5, wherein the information associated with the at least one external knowledge source includes at least one of the address, name, Uniform Resource Locator, and location of the at least one external knowledge source.

8. The first apparatus of claim 5, wherein the information relating to the amount of the external knowledge includes at least one of the following: The amount of external knowledge, and At least one criterion for controlling the amount of said external knowledge.

9. The first apparatus of claim 5, wherein the at least one strategy for the use of the external knowledge comprises at least one of the following constraints and conditions: The activation or deactivation of the external knowledge used; and Switching between different external knowledge sources.

10. The first apparatus of claim 9, wherein at least one of the constraints and conditions relates to at least one of the following: The type of communication network service, The network slice type associated with the communication network service, The location of the first device, and The type of the first device.

11. The first apparatus of claim 5, wherein the at least one strategy for the use of the external knowledge is associated with at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

12. The first apparatus of claim 5, wherein the at least one strategy for the use of the external knowledge is based on at least one of the following: At least one performance metric for the use of the external knowledge; The external knowledge shall have at least one of the following: confidence level, availability level, reliability level, and dependency level; The tasks of the communication network services; as well as The type of request input to the communication network function.

13. The first device according to any one of claims 1 to 3, wherein the at least one memory and the at least one processor further enable the first device to: Receive a report from the second device regarding the use of the external knowledge.

14. The first apparatus of claim 13, wherein the report regarding the use of the external knowledge comprises at least one of the following: Indication on whether the external knowledge is used for the communication network service; At least one external knowledge source that has already been used for the communication network services; At least one external knowledge source must be used for the communication network services; as well as At least one performance metric for the use of the external knowledge.

15. The first apparatus of claim 13, wherein the report regarding the use of the external knowledge is associated with at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

16. The first apparatus of claim 13, wherein the at least one memory and the at least one processor further enable the first apparatus to: Send to the second device a preference for the division of internal and external knowledge regarding the functions of the communication network; The preference for the division is based on the received reports on the use of the external knowledge.

17. The first apparatus of claim 16, wherein the preference for the division includes characteristics of the internal knowledge to be constructed in the communication network function.

18. The first device according to any one of claims 1 to 3, wherein the at least one memory and the at least one processor further enable the first device to: The second device receives an analysis of the relationship between the use of the external knowledge and at least one performance metric of the communication network service.

19. The first apparatus according to any one of claims 1 to 3, wherein the communication network function is implemented using a machine learning model.

20. The first apparatus of claim 19, wherein the machine learning model comprises a language-based model.

21. A second means for communication, the second means comprising: At least one processor; as well as At least one memory storing instructions that, when executed by the at least one processor, cause the second device to at least: The first device receives information relating to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside of the communication network function and the communication network function is configured to operate the communication network service. Receive a request for the communication network service from the first device; as well as In response to receiving the request, a response for the communication network service is sent to the first device. The response is based on at least one of the following: internal knowledge of the communication network functions, and external knowledge.

22. The second apparatus of claim 21, wherein the at least one memory and the at least one processor further enable the second apparatus to: Disclose information related to the ability of the communication network to support the use of the external knowledge.

23. The second apparatus of claim 22, wherein the information related to the capability of the communication network function to support the use of the external knowledge includes: The indication that the use of the external knowledge is not supported by the communication network function; or The use of the external knowledge is indicated by the network function.

24. The second apparatus according to claim 22 or 23, wherein the information related to the capability of the communication network function to support the use of the external knowledge includes at least one of the following: The form of the external knowledge. The application of the aforementioned external knowledge The storage address or region of the external knowledge. At least one entity possessing the aforementioned external knowledge. Use at least one external knowledge source's constraints or recommended description. At least one performance metric for the use of the external knowledge.

25. The second device according to any one of claims 21 to 23, wherein the information related to the use of the external knowledge includes at least one of the following: Requests for activation or deactivation of the external knowledge in relation to its use; Information related to at least one external knowledge resource to be used to retrieve the external knowledge; Information related to the amount of external knowledge to be retrieved for the service of the communication network; as well as At least one strategy used in response to the external knowledge.

26. The second apparatus of claim 25, wherein the activation or deactivation of the use of the external knowledge is requested for at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

27. The second apparatus of claim 25, wherein the information associated with the at least one external knowledge source includes at least one of the address, name, Uniform Resource Locator, and location of the at least one external knowledge source.

28. The second apparatus of claim 25, wherein the information relating to the amount of the external knowledge includes at least one of the following: The amount of external knowledge, and At least one criterion for controlling the amount of said external knowledge.

29. The second apparatus of claim 25, wherein the at least one strategy for the use of the external knowledge comprises at least one of the following constraints and conditions: The activation or deactivation of the use of the external knowledge; and Switching between different external knowledge sources.

30. The second apparatus of claim 29, wherein at least one of the constraints and conditions relates to at least one of the following: The type of communication network service, The network slice type associated with the communication network service, The location of the first device, and The type of the first device.

31. The second apparatus of claim 25, wherein the at least one strategy for the use of the external knowledge is associated with at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

32. The second apparatus of claim 25, wherein the at least one strategy for the use of the external knowledge is based on at least one of the following: At least one performance metric for the use of the external knowledge; The external knowledge shall have at least one of the following: confidence level, availability level, reliability level, and dependency level; The tasks of the communication network services; as well as The type of request input to the communication network function.

33. The second device according to any one of claims 21 to 23, wherein the at least one memory and the at least one processor further enable the second device to: Send a report to the first device regarding the use of the external knowledge.

34. The second apparatus of claim 33, wherein the report regarding the use of the external knowledge comprises at least one of the following: Indication on whether the external knowledge is used for the communication network service; At least one external knowledge source that has already been used for the communication network services; At least one external knowledge source must be used for the communication network services; as well as At least one performance metric for the use of the external knowledge.

35. The second apparatus of claim 33, wherein the report regarding the use of the external knowledge is associated with at least one of the following: The communication network functions, The session or interaction between the client of the communication network service and the functions of the communication network, and Individual requests input to the functions of the communication network.

36. The second apparatus of claim 33, wherein the at least one memory and the at least one processor further enable the second apparatus to: After sending the report on the use of the external knowledge, a preference for the division of internal and external knowledge for the communication network functions is received from the first device.

37. The second apparatus of claim 36, wherein the preference for the division includes characteristics of the internal knowledge to be constructed in the communication network function.

38. The second device according to any one of claims 21 to 23, wherein the at least one memory and the at least one processor further enable the second device to: Generate recommendations that divide internal and external knowledge based on the functions of the communication network; and The communication network functions are pre-trained or fine-tuned based on the recommendations.

39. The second device according to any one of claims 21 to 23, wherein the at least one memory and the at least one processor further enable the second device to: An analysis of the relationship between the use of the external knowledge and at least one performance metric of the communication network service is sent to the first device.

40. The second device according to any one of claims 21 to 23, wherein the at least one memory and the at least one processor further enable the second device to: Register the communication network function and related information about the ability of the communication network function to support the use of the external knowledge with a third device.

41. A method for communication, comprising: At the first device. Send information to a second device related to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside the communication network function and the communication network function is configured to operate the communication network service; Send a request for the communication network service to the second device; as well as In response to sending the request, a response for the communication network service is received from the second device, wherein the response is based on at least one of: internal knowledge of the communication network functions and the external knowledge.

42. A method for communication, comprising: At the second device, The first device receives information relating to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside of a communication network function and the communication network function is configured to operate the communication network service. Receive a request for the communication network service from the first device; as well as In response to receiving the request, a response for the communication network service is sent to the first device, wherein the response is based on at least one of the following: internal knowledge of the communication network functions and the external knowledge.

43. A first means for communication, the first means comprising: A component for sending information to a second device related to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside the communication network function and the communication network function is configured to operate the communication network service; Components for sending a request for the communication network service to the second device; as well as A component for receiving a response for the communication network service from the second device in response to sending the request. The response is based on at least one of the following: internal knowledge of the communication network functions, and external knowledge.

44. A second means for communication, the second means comprising: A component for receiving information from a first device related to the use of external knowledge for a communication network service, wherein the external knowledge is stored outside the communication network function and the communication network function is configured to operate the communication network service; Components for receiving requests for the communication network services from the first device; as well as A component for sending a response to the communication network service to the first device in response to receiving the request. The response is based on at least one of the following: internal knowledge of the communication network functions, and external knowledge.

45. A computer-readable medium comprising instructions stored thereon for causing a device to perform at least the method of claim 41 or the method of claim 42.

Images