Abstract: An apparatus comprises at least one processing device configured to generate, based on an input prompt, a first data structure comprising a textual task description associated with tasks to be performed in an environment, and to generate, based on video data of the environment, a second data structure comprising temporal dynamics information characterizing changes in spatial features of the environment over time. The at least one processing device is also configured to generate, based on images of the environment, a third data structure comprising spatial relationship information characterizing spatial relationships between objects in the environment, and to determine, utilizing at least one machine learning model that takes as input at least portions of the first, second and third data structures, actions to execute in the environment to achieve the tasks. The at least one processing device is further configured to execute the determined actions in the environment.

Abstract: A method in an illustrative embodiment includes generating, by a content generation model, first content based on a user-inputted instruction. The method further includes generating a first watermark for determining a user source based on a user identifier. The method further includes generating a second watermark for determining a model source based on a model parameter of the content generation model. The method further includes embedding the first watermark and the second watermark into the first content to generate second content with dual watermarks. In this way, a content source can be traced without modifying or retraining the content generation model, thereby simplifying the watermark embedding process, and improving the watermark embedding efficiency. In addition, the watermark can be extracted without accessing the content generation model, so that the watermark extraction process is more convenient and efficient.

Abstract: Embodiments disclosed herein include a method, an electronic device, and a computer program product for target image processing. The method includes receiving a target image and generating a first Shapley value for a feature of the target image based on the received target image. The method further includes sending, in response to satisfying a predetermined condition, a request for acquiring a second Shapley value to a cloud server. The method further includes receiving the second Shapley value for a latent feature of the target image from the cloud server, where the second Shapley value is more accurate than the first Shapley value. In some embodiments, through joint collaboration between a terminal device such as an edge device and a cloud server, rapid calculation of a Shapley value can be achieved at the terminal device, and accurate calculation of a Shapley value can be achieved at the cloud server.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating text content. The method includes determining a set containing a plurality of tokens based on a key. The method further includes based on the set, generating text content for indicating the key by a generative pre-trained model, wherein a source of the text content is verified by the text content. According to embodiments of the present disclosure, a unique watermark can be formed in the generated text content in a way that is not perceptible to a user, thereby identifying and verifying the source of the text content according to the formed watermark.

Abstract: An apparatus in an illustrative embodiment comprises at least one processing device that includes at least a processor and a memory coupled to the processor. The at least one processing device is configured to implement an artificial intelligence system comprising at least one large language model (LLM) agent, to perform in the LLM agent interactive image segmentation of at least one input image through interaction of the LLM agent with one or more users, to generate in the LLM agent an interactive image understanding comprising attention values computed by multiple distinct attention mechanisms based on one or more results of the interactive image segmentation, and to carry out additional user interactions via the LLM agent utilizing the interactive image understanding comprising the attention values computed by the multiple distinct attention mechanisms. In some embodiments, the LLM agent is illustratively utilized to provide at least a portion of an AI chatbot.

Abstract: A method in one embodiment includes acquiring a first public key and a first private key of a first user and a second public key and a second private key of a second user. The method further includes determining a first set from a glossary based on the first public key and the first private key, wherein the glossary is a set of a group of predefined tokens in a generative pre-training model. The method further includes determining a second set from the glossary based on the second public key and the second private key. The method further includes determining a set containing a plurality of tokens based on the first set and the second set. The method further includes generating text content by the generative pre-training model based on the set. According to embodiments of the present disclosure, a unique watermark can be formed in the generated text content.

Abstract: Embodiments of the present disclosure relate to a question answering method, an electronic device, and a computer program product. The method includes: determining an answer library associated with a question; determining a restricted Boltzmann machine associated with the answer library, wherein the restricted Boltzmann machine is configured to determine a set of questions that the answer library can answer and association relationships between questions in the set of questions and the answer library; and determining, using the restricted Boltzmann machine, annotation information associated with the question and targeted to the answer library. With the technical solution of the present disclosure, it is possible to determine, while determining an answer library associated with the question, annotation information that is targeted to the answer library, and to enable customer service personnel to have a more thorough understanding of the determined answer library and obtain targeted recommendation information.

Abstract: A method in an illustrative embodiment includes generating structured noise of first synthesized audio based on the first synthesized audio, and fusing a digital watermark into the structured noise. The method further includes determining a target embedding position of the digital watermark based on a spectrum of the first synthesized audio, wherein the digital watermark indicates a source of the first synthesized audio. In addition, the method further includes generating second synthesized audio based on the fused structured noise, the first synthesized audio, and the target embedding position. Through the method, not only is content of original audio preserved, but also a watermark is added, allowing a source of the audio to be recorded and traced. At the same time, the method further has a high degree of covertness, robustness, and flexibility, thereby providing a safer and more reliable environment for the synthesized audio, and improving the user experience.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a program product for a large language model (LLM). The method includes: receiving a user input for an LLM agent; determining role information and user-related alignment information for the LLM agent based on the user input; generating a prompt including the role information and the alignment information; and generating an answer to the user input by providing the prompt to the LLM. In this way, appropriate role and user-related alignment information can be configured for the LLM agent to help the LLM agent to provide a desired answer for users in open application fields, thereby improving user experience.

Abstract: A method in one embodiment includes acquiring a first public key and a first private key of a first user and a second public key and a second private key of a second user. The method further includes determining a first set from a glossary based on the first public key and the first private key, wherein the glossary is a set of a group of predefined tokens in a generative pre-training model. The method further includes determining a second set from the glossary based on the second public key and the second private key. The method further includes determining a set containing a plurality of tokens based on the first set and the second set. The method further includes generating text content by the generative pre-training model based on the set. According to embodiments of the present disclosure, a unique watermark can be formed in the generated text content.

Abstract: A method in an illustrative embodiment includes generating, by a content generation model, first content based on a user-inputted instruction. The method further includes generating a first watermark for determining a user source based on a user identifier. The method further includes generating a second watermark for determining a model source based on a model parameter of the content generation model. The method further includes embedding the first watermark and the second watermark into the first content to generate second content with dual watermarks. In this way, a content source can be traced without modifying or retraining the content generation model, thereby simplifying the watermark embedding process, and improving the watermark embedding efficiency. In addition, the watermark can be extracted without accessing the content generation model, so that the watermark extraction process is more convenient and efficient.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating text content. The method includes determining a set containing a plurality of tokens based on a key. The method further includes based on the set, generating text content for indicating the key by a generative pre-trained model, wherein a source of the text content is verified by the text content. According to embodiments of the present disclosure, a unique watermark can be formed in the generated text content in a way that is not perceptible to a user, thereby identifying and verifying the source of the text content according to the formed watermark.

Abstract: The present disclosure relates to a method, a device, and a computer program product for generating a response about a document. The method includes: according to a first document, retrieving multiple second documents associated with the first document; determining duplication between the first document and the multiple second documents through contextual analysis of the first document and the multiple second documents; determining gaps between the first document and the multiple second documents through causal analysis of the first document and the multiple second documents; and generating a response about the first document according to the duplication and the gaps using a language model. In this way, by considering both the context and the causality in various documents, it is possible to conduct duplication and gap detection for more complex documents, and thus generate a more accurate response about the document.

Abstract: Embodiments of the present disclosure relate to a method for wireless communication, a wireless port connection device, and a computer program product. The method in one embodiment is illustratively performed by a first wireless port device. The method includes: sending a first public key of the first wireless port device to an authentication server; receiving a signed first certificate from the authentication server, the first certificate comprising a server public key of the authentication server; receiving a signed second certificate from a second wireless port connection device; acquiring a second public key of the second wireless port connection device in the second certificate using the server public key; generating a symmetric key using the second public key; encrypting data using the symmetric key; and sending the encrypted data to the second wireless port connection device.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product that involve address encryption. The method includes: in response to receiving a packet from a source device, extracting a first address representing an address of the source device from the packet. The method further includes: mixing the first address with a random number to generate a mixed address, wherein the number of bits of the first address is the same as the number of bits of the random number. The method further includes: encrypting the mixed address. Moreover, the method further includes: generating a second address representing the address of the source device at least partially based on the encrypted mixed address, wherein the second address includes a prefix field and an interface identifier field. Illustrative embodiments can improve the security of the source device, and reduce computational overhead during address encryption and decryption.

Abstract: According to an embodiment, a performance benchmark database is obtained, where the performance benchmark database at least includes structural data of one or more deep neural network models, time performance data and computing resource consumption data of a plurality of deep learning applications based on the one or more deep neural network models; a training dataset is extracted based on the performance benchmark database, where the training dataset has a plurality of parameter dimensions, the plurality of parameter dimensions including: structures of deep neural network models of the plurality of deep learning applications, resource configuration of the plurality of deep learning applications, and training time of the plurality of deep learning applications; and correspondence among the parameter dimensions of the training dataset is created so as to create an estimation model for estimating resources utilized by deep learning applications.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for virtual reality modeling. The method includes obtaining a first image of a real object at a first viewing angle. The method further includes obtaining, based on the first image, an initial three-dimensional model corresponding to the real object. The method further includes determining a second image of the real object at a second viewing angle different from the first viewing angle by using the initial three-dimensional model. The second image has texture characteristics of the real object. The method further includes generating a target three-dimensional model used for virtual reality and corresponding to the real object by using the second image and the first image. Through the solution, a three-dimensional model which can reproduce a real object with high quality may be obtained based on a small number of input images.

Abstract: The present disclosure relates to a method, a device, and a computer program product for image processing. The method includes acquiring a plurality of images and a first video, wherein the plurality of images indicate visual information of a source object from a plurality of perspectives, and the first video indicates animation of a target object. The method further includes generating a three-dimensional model for the source object based on the plurality of images, and generating a plurality of animation models for the target object based on the first video. The method further includes fusing the three-dimensional model for the source object and the plurality of animation models for the target object to generate a second video for the source object, wherein in the second video, the target object in the first video is replaced with the source object.

Abstract: Embodiments of the present disclosure relate to a method, an apparatus, a device, and a medium for protecting an internet-of-things device. The method includes receiving a data packet associated with the internet-of-things device. The method further includes determining whether the data packet is for a to-be-protected target internet-of-things device based on a service identification in the data packet. The method further includes filtering, in response to the data packet being for the to-be-protected target internet-of-things device, the data packet based on service forwarding information related to a service of the target internet-of-things device to protect the target internet-of-things device. Through the method, normal packets are forwarded and possible attack data packets are abandoned by filtering the data packets, and thus the internet-of-things device is prevented from being attacked, which improves the security of the internet-of-things device and improves the user experience.

Abstract: Embodiments of the present disclosure relate to a task processing method, a task processing device, and a computer program product. The method includes: receiving a task to be processed by a first neural network at a first computing node; determining that the first neural network failed to process the task; and sending the task to a second computing node for processing by a second neural network at the second computing node, wherein compared with the first neural network, the second neural network has at least one of the following: higher processing consumption and higher processing performance. With the technical solution of the present disclosure, a task can be processed quickly and accurately using a neural network with low processing consumption.