Abstract: A data transmission method, a data transmission apparatus, and a data transmission device are provided, where the method on the terminal side includes: obtaining, by the terminal, an ACK/NACK sequence to be fed back; selecting, by the terminal, a machine learning model applied to the ACK/NACK sequence; compressing, by the terminal, the ACK/NACK sequence through the machine learning model to obtain an indication sequence; and transmitting, by the terminal, the indication sequence to the network device.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated computer code. Lines of code can be masked from a candidate code to obtain perturbed codes. Missing code can be generated from the perturbed codes by employing an AI code generator model to obtain machine-filled codes. Probabilities of the candidate code probability and the machine-filled codes as AI-generated can be predicted by employing a surrogate model. The candidate code can be distinguished as AI-generated by comparing the probabilities against a detection threshold to obtain detection results.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated text. A candidate text can be truncated to obtain a prefix text and a remainder text by employing a truncation module. Regenerated model texts can be regenerated by utilizing the prefix text by employing an AI text generation model. Detection results can be predicted by comparing n-gram similarities of the regenerated model texts and the remainder text. The candidate text can be distinguished as AI generated text by providing explanation texts based on the detection results.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated computer code. Lines of code can be masked from a candidate code to obtain perturbed codes. Missing code can be generated from the perturbed codes by employing an AI code generator model to obtain machine-filled codes. Probabilities of the candidate code probability and the machine-filled codes as AI-generated can be predicted by employing a surrogate model. The candidate code can be distinguished as AI-generated by comparing the probabilities against a detection threshold to obtain detection results.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated computer code. Lines of code can be masked from a candidate code to obtain perturbed codes. Missing code can be generated from the perturbed codes by employing an AI code generator model to obtain machine-filled codes. Probabilities of the candidate code probability and the machine-filled codes as AI-generated can be predicted by employing a surrogate model. The candidate code can be distinguished as AI-generated by comparing the probabilities against a detection threshold to obtain detection results.

Abstract: The present disclosure, which relates to the field of communication technologies, provides a method and a device for codebook determination, a method and a device for signal processing, and a readable storage medium, to improve the effect of suppressing beam squint and reduce the complexity of codebook design. The method includes: dividing a broadband signal into multiple sub-bands; determining a first analog beamforming matrix corresponding to each sub-band; determining a first digital precoding matrix corresponding to each first analog beamforming matrix; and forming a transmission codebook based on each first digital precoding matrix and a first sub-band-related parameter for each sub-band; where a bandwidth of each sub-band meets following requirement: a difference between a first direction angle and a second direction angle is less than a preset threshold.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated computer code. Lines of code can be masked from a candidate code to obtain perturbed codes. Missing code can be generated from the perturbed codes by employing an AI code generator model to obtain machine-filled codes. Probabilities of the candidate code probability and the machine-filled codes as AI-generated can be predicted by employing a surrogate model. The candidate code can be distinguished as AI-generated by comparing the probabilities against a detection threshold to obtain detection results.

Abstract: Systems and methods for detecting Large Language Model (LLM) generated text are provided. The systems and methods include sampling a text passage to generate alternative samples conditioned on the text passage based on a next token prediction in a surrogate LLM model and scoring a likelihood that the test passage sample is generated by an LLM model. The scoring includes a conditional probability which quantifies a distribution gap of a log of logits from the surrogate LLM model. The systems and methods further include comparing the scored text passage with a sample text generated in the surrogate LLM model trained to imitate a target LLM model. The comparison includes transforming the scores into a scaled representation and normalizing the scores.

Abstract: A terminal equipment includes: a receiver configured to: receive a first signal including a first parameter configuring or indicating one or more repetition, and receive a second signal including downlink control information indicating a first repetition or a second repetition, the first repetition being a repetition of slot level, the second repetition being a repetition of symbol level; and a transmitter configured to transmit, according to the first parameter and downlink control information, one or more actual repetitions corresponding to the same transport block in one or more time-domain resource segments in which the one or more actual repetitions are located, each of the one or more actual repetitions comprising a data channel and a reference signal.

Abstract: The present disclosure provides a near-field beam search method and apparatus, and a storage medium, in which a base station determines a grid where a terminal is located from at least one type of preset grid of a near-field coverage range of the base station; the base station determines a target transmitting beam codebook according to the grid where the terminal is located and a base station transmitting beam codebook set corresponding to the preset grid.

Abstract: A near-field beam search method, including: a base station determines a grid where a terminal is located from at least one type of preset grid of a near-field coverage range of the base station; the base station determines a target transmitting beam codebook according to the grid and a base station transmitting beam codebook set corresponding to the preset grid, where the base station transmitting beam codebook set includes a transmitting beam codebook corresponding to each grid in the preset grid; the base station sends information of the target transmission beam codebook to the terminal, and the terminal determines a target receiving beam codebook from a terminal receiving beam codebook set corresponding to the preset grid according to the information; the base station sends a signal to the terminal by using the target transmitting beam codebook, and the terminal receives the signal by using the target receiving beam codebook.

Abstract: A method and apparatus for transmitting a reference signal and a communication system. The reference signal transmission method includes: a transmitter end generates one or more signals corresponding to the same transport block, the one or more signals including a data channel and a reference signal; and the transmitter end transmits the one or more signals in one or more time-domain resource segments, positions of reference signals being determined based on the one or more time-domain resource segments. The terminal equipment and the network device may be able to determine the positions of the reference signals in an actual repetition according to time-domain resource segments occupied by the actual repetition.

Abstract: Methods and apparatuses for transmitting and receiving downlink control information and a communication system. The apparatus for transmitting downlink control information (DCI) includes: a first transmitter configured to transmit to a terminal equipment first DCI indicating an interrupted first time-frequency resource used for transmitting uplink data, the first DCI at least comprising a first field indicating a position of the first time-frequency resource.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated computer code. Lines of code can be masked from a candidate code to obtain perturbed codes. Missing code can be generated from the perturbed codes by employing an AI code generator model to obtain machine-filled codes. Probabilities of the candidate code probability and the machine-filled codes as AI-generated can be predicted by employing a surrogate model. The candidate code can be distinguished as AI-generated by comparing the probabilities against a detection threshold to obtain detection results.

Abstract: Systems and methods are provided for dynamic prompt tuning in image processing, including decomposing a received image into segments sized to balance detail retention and computational efficiency for processing by an embedding algorithm designed for token generation, generating tokenized image data by transforming each of the decomposed segments into a sequence of tokens using an embedding process that includes a convolutional neural network, and dynamically computing parameters for inserting prompts into the sequence of tokens, including a position and length of the prompts, utilizing a one-layer neural network combined with a continuous relaxation of a discrete distribution for optimizing categorical decision-making. Soft prompts are created based on the dynamically computed parameters and the soft prompts are integrated with the tokenized image data. The integrated image data and prompts are processed using a pretrained vision model with a frozen backbone to enhance image feature recognition.

Abstract: Systems and methods for detecting artificial intelligence (AI) generated text. A candidate text can be truncated to obtain a prefix text and a remainder text by employing a truncation module. Regenerated model texts can be regenerated by utilizing the prefix text by employing an AI text generation model. Detection results can be predicted by comparing n-gram similarities of the regenerated model texts and the remainder text. The candidate text can be distinguished as AI generated text by providing explanation texts based on the detection results.

Abstract: The present application provides a UE capability updating method and apparatus, and a device. The method comprises: updating a machine learning model used for performing communication data processing; processing communication data according to the updated machine learning model, and determining at least one performance index related to a UE capability; when determining, according to the at least one performance index, that the UE capability is updated, determining the updated UE capability information; and reporting the updated UE capability information to a network. According to the method provided by the present application, introducing a method for identifying a UE capability change and reporting the updated UE capability solves the problem that after the UE capability is upgraded, a resource is wasted or a service requirement for the upgrading of a UE cannot be satisfied caused due to the fact that a network device still performs scheduling according to an original UE capability.

Abstract: A beam failure recovery method and device and a communication system. The device includes: a receiver configured to, after a period of time upon the terminal equipment receive a beam failure recovery (BFR) response and before activation or reconfiguration signaling, monitor and/or receive a downlink signal in a first cell of the terminal equipment according to an antenna quasi-co-located (QCL) parameter identical to a selected reference signal (q_new) or a reference signal (q_new) indicated by a higher layer.

Abstract: Provided is a model update method in a communication system, including: determining that a communication environment changes, and finding, according to a first mapping relationship set, a machine learning model suitable for a current communication environment from a plurality of machine learning models associated with the first mapping relationship set, to form a first machine learning model set; and deploying the machine learning model of the first machine learning model set on the access network device, and/or transmitting first information to a terminal, where the first information is used to instruct the terminal to deploy the machine learning model of the first machine learning model set.

Abstract: Methods and systems for prompt tuning include training a tuning function to set prompt position, prompt length, or prompt pool based on a language processing task. The tuning function is applied to an input query to generate a combined input, with prompt text having the prompt length, being selected according to the prompt pool, and being added to the input query at the prompt position. The combined input is applied to a language model.