Abstract: A system may be configured for implementing targeted attacks on deep reinforcement learning-based autonomous driving with learned visual patterns. In some examples, processing circuitry receives first input specifying an initial state for a driving environment and user configurable input specifying a target state. Processing circuitry may generate a representative dataset of the driving environment by performing multiple rollouts of the vehicle through the driving environment, including performing an action for the vehicle from the initial state with variable strength noise added to determine a next state for each rollout resulting from the action. Processing circuitry may train an artificial intelligence model to output a next predicted state based on the representative dataset as training input. In such an example, processing circuitry outputs from the artificial intelligence model, an attack plan against the autonomous driving agent to achieve the target state from the initial state.

Abstract: The present invention belongs to the field of biotechnology and provides the application of the watermelon flesh color intensity control gene CIFCI gene and its tandem repeat sequences in promoter region in the identification of watermelon flesh color intensity. Based on the presence of the tandem repeat region in the promoter of the CIFCI gene in watermelon varieties, the present invention designs a PCR marker for convenient detection, which can quickly detect the possible flesh color phenotypes in the hybrid offspring with light and deep flesh colors at the seedling stage. The application of this gene marker in marker-assisted breeding can improve the quality of watermelon fruits.

Abstract: The present invention provides a general salt-assisted chemical vapor deposition (SA-CVD) synthetic method for the high-yield preparation of 1D hexagonal-phase MxV6S8(M=K, Rb, Cs) and KxV6SySe8-y nanowires. The resulting nanowires exhibit typical metallic properties, which can be used as a good van der Waals contact for achieving high-performance fermi level pinning free transistors. The present synthesis method allows a more systematic investigation of the intrinsic properties of hexagonal-phase V6S8 structures.

Abstract: Embodiments of the disclosue relate to a method, apparatus, device, and storage medium of training a music compression system. The method includes: obtaining a first encoded representation associated with training music content; processing the first encoded representation with the discrete encoder to generate a first set of discrete features corresponding to first music data and a second set of discrete features corresponding to second music data; decoding the first set of discrete features with the first discrete decoder to obtain a first audio feature corresponding to the first music data, and decoding the second set of discrete features with the second discrete decoder to obtain a second audio feature corresponding to the second music data; and determining a training loss based on the first audio feature, the second audio feature and the training music content, and adjusting parameters of the discrete encoder and the discrete decoders based on training loss.

Abstract: A water purification bottle includes an inner bottle body and an outer bottle body, wherein a filter unit is detachably arranged at the bottom end of the inner bottle body; a periphery of the inner bottle body is actively matched with an inner perimeter of the outer bottle body; when the outer bottle body is filled with a mixture of liquid and solid, only the liquid of the mixture runs through the filter unit, and flows into the inner bottle body.

Abstract: A system and associated methods for decentralized attribution of GAN models is disclosed. Given a group of models derived from the same dataset and published by different users, attributability is achieved when a public verification service associated with each model (a linear classifier) returns positive only for outputs of that model. Each model is parameterized by keys distributed by a registry. The keys are computed from first-order sufficient conditions for decentralized attribution. The keys are orthogonal or opposite to each other and belong to a subspace dependent on the data distribution and the architecture of the generative model.

Abstract: A bilateral teleoperation system includes: a primary-end operation platform and a secondary-end operation platform. The primary-end operation platform includes: a primary-end support, primary-end mechanical arms, a mechanical hand control assembly, and a first controller, a root end of the primary-end mechanical arm being arranged on the primary-end support, and a tail end of the primary-end mechanical arm being connected to the mechanical hand control assembly.

Abstract: A volume adjustment method, a device and a storage medium are provided. The method includes: determining a reference volume corresponding to a target audio to be played; inputting target audio characteristic information corresponding to the target audio, target user group characteristic information provided by a target user group, current scenario characteristic information and the reference volume to an adjustment information prediction model that is pre-trained, to predict volume adjustment information and obtain target volume adjustment information; determining a target volume corresponding to the target audio according to the target volume adjustment information and the reference volume; and performing volume adjustment on the target audio based on the target volume, to enable an adjusted target audio to be played at the target volume.

Abstract: Disclosed are a new crystal form of treprostinil sodium salt (the structural formula thereof is shown as formula I) and a preparation method therefor. Specifically, disclosed are a five and half hydrates of the treprostinil sodium salt and some dehydrated crystal forms thereof. The new crystal form has better stability and hygroscopicity. As a treprostinil crude drug, the new crystal form can meet the requirements for storage and loading and transportation, is stable in terms of quality and can guarantee the controllable quality of subsequent preparation products.

Abstract: Examples systems and methods for improved artificial intelligence generative model image attribution include a processor accessing instructions stored on a machine-readable storage medium. The instructions can be executable by the processor to obtain latent semantic dimensions by applying a principal component analysis on a latent distribution. Perturbations can be applied along the latent semantic dimensions according to a user-specific key and a predetermined perturbations strength to embed a model-specific fingerprint to embed a fingerprinted latent variable directly into the generative model such that the generative model generates images with model-specific fingerprints.

Abstract: Certain aspects of the present disclosure provide techniques for expediting mobility. One example method generally involves receiving signaling, from a serving cell of a first radio access technology (RAT), indicating a first list of frequencies associated with the first RAT and corresponding priority levels: performing idle measurements for frequencies in the first list, while camped on a serving cell of a second RAT: setting cell reselection thresholds to apply to the idle measurements for the frequencies in the first list, based on a relative priority of the serving cell of the second RAT to the frequencies in the first list: and performing cell reselection procedures based on the idle measurements for the frequencies and the determined cell reselection thresholds.

Abstract: Embodiments of the disclosure provide a solution for speech generation. A method includes: determining, based on a target text, a plurality of phoneme feature representations corresponding to a sequence of phonemes in the target text and respective phoneme durations for the plurality of phoneme feature representations; extending the plurality of phoneme feature representations based on the respective phoneme durations, to obtain an extended sequence of phoneme feature representations; masking at least one phoneme feature representation in the extended sequence of phoneme feature representations, to obtain a sequence of masked phoneme feature representations; and generating a target speech corresponding to the target text at least based on the sequence of masked phoneme feature representations.

Abstract: The present invention provides a general salt-assisted chemical vapor deposition (SA-CVD) synthetic method for the high-yield preparation of 1D hexagonal-phase MxV6S8(M=K, Rb, Cs) and KxV6SySe8-y nanowires. The resulting nanowires exhibit typical metallic properties, which can be used as a good van der Waals contact for achieving high-performance fermi level pinning free transistors. The present synthesis method allows a more systematic investigation of the intrinsic properties of hexagonal-phase V6S8 structures.

Abstract: Embodiments of the disclosure provide a solution for model training. A method includes: obtaining a training sample for training a machine learning model, the training sample comprising a sample speech, a sample text corresponding to the sample speech and speech duration information for the sample text, the machine learning model being configured to perform a plurality of tasks for speech processing; extracting a speech feature representation from the sample speech, a text feature representation from the sample text, a phoneme feature representation and a duration feature representation for the phoneme feature representation from the speech duration information; and training, according to the plurality of tasks, the machine learning model based on at least one of: the speech feature representation, the text feature representation or a combination of the phoneme feature representation and the duration feature representation.

Abstract: Methods, systems, and devices for wireless communications are described. Generally, a UE in a connected state with a cell of a first radio access technology (RAT) type may trigger a UE initiated inter-RAT redirection procedure to a neighboring cell of a different, higher priority level, RAT type if the network does not configure the UE to perform connected measurements for candidate neighbor cells of the higher priority level RAT type. A UE may receive an indication of candidate neighbor cells of a higher priority level RAT type. The UE may perform UE initiated measurements on the indicated candidate neighbor cells of the higher priority level RAT type. The UE may trigger an inter-RAT redirection procedure to one of the neighbor cells of the higher priority level RAT if the UE finds a suitable cell based on the UE initiated measurements.

Abstract: Embodiments of the present disclosure provide a task processing method and apparatus, an electronic device, and a storage medium. The method includes: sending first notification information to a client of at least one task to be delivered that is associated with a current preset location, where the first notification information includes a wait time; receiving first feedback information from the at least one client, and determining a suspended task for deferred processing based on the first feedback information; and in response to receiving a task processing request sent by a target client corresponding to a target suspended task, determining a target processing manner corresponding to the target suspended task based on a current location, to process the target suspended task based on the target processing manner.

Abstract: Embodiments of the present disclosure provide a robot dispatching method and apparatus, an electronic device, and a storage medium. The method includes: displaying at least one dispatchable robot available for switching in response to a robot switching instruction; and determining a target robot in response to a trigger operation for the at least one dispatchable robot, and sending robot information of the target robot to a dispatching system, so that the dispatching system associates a pending task associated with an abnormal robot with the target robot. According to the technical solutions of the embodiments of the present disclosure, an effect of switching to another robot to resume an associated task based on one-click operation is achieved, which in turn improves the universality and efficiency of a robot switching operation.

Abstract: Embodiments of the present disclosure provide a method and apparatus, an electronic device, and a storage medium for task processing. The method is applied to a robot and includes: displaying at least one display control and an abnormality reporting control on a display screen of the robot, where the display control is configured to display task-associated information of a task to be processed; displaying an abnormal task selection page in response to a trigger operation for the abnormality reporting control, to determine a target abnormal task based on a trigger operation on the abnormal task selection page, where the abnormal task selection page includes the at least one display control; and sending task-associated information and abnormality information of the target abnormal task to a target terminal in response to a trigger operation of uploading the target abnormal task.

Abstract: Methods and apparatuses for recognizing a two-dimensional code are disclosed. In an implementation, a method comprises: identifying, by an electronic device, a two-dimensional code, wherein the two-dimensional code comprises an image region in the center and a ring region surrounding the image region, wherein the ring region comprises a code region including a first code region and a second code region, a first spacing region, and a second spacing region, wherein the first spacing region and the second spacing region are arranged between the first code region and the second code region, determining values corresponding to a plurality of code elements in the code region, and recognizing, by the electronic device based on the values, first information corresponding to the two-dimensional code.

Abstract: Provided are dynamic sealing structure and rotary kiln apparatus. Dynamic sealing structure includes kiln tail, discharging cover, sealing mechanism, guiding mechanism and balancing device. Discharging cover includes discharging cover cylinder and discharging cover end face connected to same, discharging cover cylinder is arranged on outer side of circumferential wall of kiln tail in sleeving manner, and discharging cover end face and end face of the kiln tail are arranged at intervals. Sealing mechanism is arranged between discharging cover cylinder and kiln tail and is connected to inner wall of discharging cover cylinder. Guiding mechanism is arranged between inner wall of discharging cover cylinder and outer side of kiln tail, and is located on one side or two sides of sealing mechanism in axial direction of rotary kiln, rotating gap is provided between guiding mechanism and kiln tail, and balancing device is connected to outer side of the discharging cover cylinder.