MULTI-RESOLUTION FIELD REPRESENTATIONS IN NEURAL NETWORKS

    公开(公告)号:US20250094780A1

    公开(公告)日:2025-03-20

    申请号:US18468203

    申请日:2023-09-15

    Abstract: Certain aspects provide techniques and apparatuses for efficiently processing inputs in a neural network using multiple receptive field sizes. An example method includes partitioning a first input into a first set of channels and a second set of channels. At a first layer of a neural network, the first set of channels and the second set of channels are convolved into a first output having a smaller dimensionality a dimensionality of the first input. The first set of channels and the first output are concatenated into a second input. The second input is convolved into a second output via a second layer of the neural network, wherein the second output merges a first receptive field generated by the first layer with a larger second receptive field generated by the second layer. One or more actions are taken based on at least one of the first output and the second output.

    HYBRID GENERATIVE ARTIFICIAL INTELLIGENCE MODELS

    公开(公告)号:US20240362468A1

    公开(公告)日:2024-10-31

    申请号:US18543533

    申请日:2023-12-18

    CPC classification number: G06N3/0475

    Abstract: Certain aspects of the present disclosure provide techniques and apparatus for generating a response to an input query using a generative artificial intelligence model. An example method generally includes receiving an input for processing. A prompt representing the received input is generated based on the received input, contextual information associated with the received prompt, and a prompt-generating artificial intelligence model. The generated prompt is output to a generative artificial intelligence model for processing. A response to the generated prompt is received from the generative artificial intelligence model and output as a response to the received input.

    SPECULATIVE DECODING IN AUTOREGRESSIVE GENERATIVE ARTIFICIAL INTELLIGENCE MODELS

    公开(公告)号:US20240320433A1

    公开(公告)日:2024-09-26

    申请号:US18479672

    申请日:2023-10-02

    CPC classification number: G06F40/284 G06F16/2246

    Abstract: Certain aspects of the present disclosure provide techniques and apparatus for generating a response to an input query using generative models. The method generally includes generating, based on an input query and a first generative model, a first plurality of sets of tokens. The first plurality of sets of tokens are output to a second generative model for verification. While waiting to receive an indication of a selected set of tokens from the first plurality of sets of tokens, a second plurality of sets of tokens are speculatively generated. The indication of a selected set of tokens from the first plurality of sets of tokens is received. Tokens from the second plurality of sets of tokens associated with the selected set of tokens are output to the second generative model for verification, and the selected set of tokens is output as a response to the input query.

    SIGNALING FOR SLOT AGGREGATION
    8.
    发明公开

    公开(公告)号:US20230345434A1

    公开(公告)日:2023-10-26

    申请号:US18343918

    申请日:2023-06-29

    Abstract: Certain aspects of the present disclosure provide techniques for slot aggregation in new radio (NR) systems. A method of wireless communication by a user equipment (UE) includes receiving radio resource control (RRC) signaling providing a semi-static configuration for transmission and/or reception of a repeated transport block (TB) and/or a different TB in each of a plurality of aggregated slots. The UE transmits or receives the TBs in the plurality of aggregated slots based on the semi-static configuration. Another method is provided in which the UE transmits and/or receives a demodulation reference signal (DMRS) in each of a plurality of aggregated slots, each DMRS associated with a repeated TB or a different TB in the slot. The UE determines, based on received signaling, whether the DMRS use a same precoder or a different precoder and modulates and/or demodulates the TBs in the plurality of aggregated slots based on the determination.

    POSITIONING REFERENCE SIGNAL STAGGERING CONFIGURATION

    公开(公告)号:US20220417896A1

    公开(公告)日:2022-12-29

    申请号:US17930037

    申请日:2022-09-06

    Abstract: Disclosed are methods and apparatuses for resolving aliasing ambiguities produced when using channel state information reference signal, a sounding reference signal (SRS) or other transmission as a positioning reference signal (PRS) occupying a subset of tones of a PRS bandwidth. The aliasing ambiguity results in a plurality of different possible positioning measurements, such as time of arrival (TOA), reference signal timing difference (RSTD), or reception to transmission difference (Rx−Tx). The aliasing ambiguity may be resolved using a previous position estimate that may be used to produce an approximation of the expected positioning measurement. A position estimate may be generated using a multiple stage PRS configuration, in which one or more stages provide a coarse position estimate that has no ambiguity, which can be used to resolve the ambiguity of a more accurate position estimate resulting from the PRS signal.

Patent Agency Ranking