Abstract: Embodiments described herein provide a large language model (LLM) based AI agent that adopts Monte-Carlo Tree Search (MCTS) to execute a task. The LLM is prompted with a task description and it responds with its first attempted list of actions. Based on the success or failure of the first attempt, the LLM is prompted with an updated prompt which includes feedback from the first attempt based on a determined reward. The prompt may include a relative “score” for each action taken at each step. A numeric score may be mapped to a set of pre-defined text labels, such as “high” or “low” value putting the score in a form more suited for an LLM prompt. In this way, the LLM is iteratively given prompts which are updated with the scores from each action taken at each previous iterations so that it traverses different paths on the tree in each iteration.

Abstract: The invention relates to an Automatic garage door control system and method based on ats timetable, which are used for implementing remote automatic control of garage doors. The ATS timetable-based automatic control system comprises a garage door plan management module, a garage door information maintenance module and a garage door control processing module, wherein the garage door control processing module outputs garage door control commands and alarm events according to times calculated by the garage door plan management module and garage door basic information provided by the garage door information maintenance module. Compared with the prior art, the invention realizes automatic processing, provides multiple options, and is good in timeliness and practicability.

Abstract: Aspects present herein relate to methods and devices for wireless communication including an apparatus, e.g., a UE and/or a base station. The apparatus may identify at least one scheduling offset associated with a propagation delay between a base station of a NTN and the UE. The apparatus may also select a first scheduling offset of the at least one scheduling offset associated with the propagation delay between the base station and the UE. Additionally, the apparatus may transmit, to the base station, an uplink transmission based on the first scheduling offset, the uplink transmission being associated with a PUSCH, a PUCCH, or a PRACH.

Abstract: A scanning apparatus (110), a medical image obtaining method, and a medical image obtaining system (100) are provided. The scanning apparatus (110) comprises a gantry, a controller, a C-shaped arm, a radiation source (112), and a detector (113). The radiation source (112) and the detector (113) are arranged at both ends of the C-shaped arm. The C-shaped arm is connected to the gantry. The controller is configured to control a motion of the gantry to drive the C-shaped arm to move so as to scan a target object.

Abstract: Provided is a method and a system for evaluating enzymatic reaction feasibility based on multiple tasks and molecular multi-modal features. An enzymatic reaction feasibility dataset is constructed with a public dataset and a bioengineering reaction rule template library; SMILES sequence features and Morgan fingerprint spatial structure features of a product molecule and a substrate molecule of a reaction are used as inputs to a neural network; a dual-branch network is constructed based on an attention mechanism and a convolutional neural network to extract molecular multi-modal features; a product SMILES sequence generation task is taken as a secondary task to strengthen the capability of a model learning sequence feature; richer features are provided for an enzymatic reaction feasibility evaluation task; and the trained model is effectively enabled to accurately determine reaction feasibility by taking the molecular multi-modal features into overall consideration.

Abstract: Methods, systems, apparatuses, devices, and computer program products are described. A flow generation service may receive a natural language input that indicates instructions for automating a task according to a first process flow. Using a large language model (LLM), the flow generation service may decompose the natural language input into a set of elements (e.g., logical actions) and connectors, where the LLM may be trained on first metadata corresponding to a second process flow that is created manually by a user. In addition, using the LLM, the flow generation service may generate second metadata corresponding to each of the set of elements based on decomposing the natural language input. The flow generation service may sequence and merge the set of elements to generate the first process flow. In some examples, the flow generation service may send, for display to a user interface of a user device, the first process flow.

Abstract: A mammography system and a method for imaging at least part of a breast with the mammography system are disclosed. The method includes: obtaining pre-exposure images respectively corresponding to a plurality of phantoms by performing pre-exposure for each of the plurality of phantoms with a respective predetermined dosage, each phantom having its respective compression thickness and component composition; determining a relationship among a target grayscale, the compression thickness and the component composition by performing de-scattering processing on each pre-exposure image based on the compression thickness and the component composition of the phantom corresponding to the pre-exposure image; determining a target dosage for a target breast based on a compression thickness and a component composition of the target breast, and the relationship; and acquiring a main exposure image by performing exposure for the target breast based on the target dosage.

Abstract: Embodiments of the present disclosure provide a method and a system for determining an image control point. The method may include obtaining a first image; determining a plurality of initial control points in the first image; dividing the first image into at least two sub-regions; and determining, based on one or more features of the plurality of initial control points, one or more target control points in each of the at least two sub-regions.

Abstract: Embodiments described herein provide a method of predicting an action by a plurality of language model augmented agents (LAAs). In at least one embodiment, a controller receives a task instruction to be performed using an environment. The controller receives an observation of a first state from the environment. The controller selects a LAA from the plurality of LAAs based on the task instruction and the observation. The controller obtains an output from the selected LAA generated using an input combining the task instruction, the observation, and an LAA-specific prompt template. The controller determines the action based on the output. The controller causes the action to be performed on the environment thereby causing the first state of the environment to change to a second state.

Abstract: Embodiments described herein provide for optimizing a language model (LM) agent. In at least one embodiment, and LM agent comprises an “actor” LM and a “retrospective LM which provides reflections on attempts by the actor LM. The reflections are used to update subsequent prompts to the actor LM. Optimizing the LM agent comprises fine-tuning parameters of the retrospective LM while keeping parameters of the actor LM frozen. A gradient may be determined by a change in reward from the environment based on actions taken by the actor LM with and without a reflection of the retrospective LM. Using this gradient, parameters of the retrospective LM may be updated via backpropagation.

Abstract: The present disclosure is related to systems and methods for processing X-ray images. A method for processing an X-ray image may include obtaining an X-ray image; and determining a metal image based on the X-ray image by using a trained metal detection model. The metal image includes information of a metal object in the X-ray image.

Abstract: The present disclosure relates to systems and methods for image processing. The system may obtain at least one image of an object. For each of the at least one image, the system may determine a recognition result of the image. The recognition result may include an image type of the image, a type of a lesion in the image, a region of the lesion in the image, and/or an image feature of the image. Further, the system may process the at least one image of the object based on at least one recognition result corresponding to the at least one image.

Abstract: Methods, systems, and devices for wireless communication are described. Wireless communications systems may support uplink random access channel (RACH) transmissions on multiple beams and over multiple component carriers (CCs). A wireless device may transmit a random access preamble to a base station in a first RACH transmission, which may indicate a set of CCs over which the base station may respond with a random access response (RAR) in a second RACH transmission. The second RACH transmission may then include an indication for which CCs the wireless device may use for a subsequent RACH transmission (e.g., a RACH message 3). The wireless device may also indicate a beam index and/or time-frequency resources associated with beams and/or CCs used for such cell acquisition transmissions. In other examples, the base station may indicate resources (e.g., via a handover command or RACH command) for wireless device scheduling request and/or beam-failure recovery request transmission.

Abstract: A two-dimensional porous silicon, and a preparation method and use thereof in lithium ion batteries. A silicocalcium powder is soaked in a hydrochloric acid solution, filtered under suction after reaction, and dried to obtain a siloxene powder. The siloxene powder is thermally treated in a vacuum environment, in which siloxene is oxidized into SiOx during the heating process, and SiOx is disproportionated in further high-temperature treatment to produce uniformly distributed two-dimensional Si/SiO2. The two-dimensional Si/SiO2 is soaked in a hydrofluoric acid solution, in which hydrofluoric acid reacts with SiO2. After complete reaction, a two-dimensional silicon material having a porous structure is obtained after repeatedly centrifugation and washing, and drying under vacuum.

Abstract: Systems and methods for medical image reconstruction are provided. The systems may determine a target temporal resolution of a target scan on a target subject within a scanning angle range. The system may determine, based on the target temporal resolution, a plurality of reconstruction angle ranges each of which being within the scanning angle range. The system may obtain scan data acquired by causing an imaging device to perform the target scan on the target subject within the scanning angle range. The system may generate a plurality of reconstruction images corresponding to the plurality of reconstruction angle ranges based on the scan data.

Abstract: A method may include obtaining an image of a subject and determining a plurality of image blocks in the image. The method may also include extracting grayscale features from each of the plurality of image blocks and determining, based on the grayscale features, a segmentation threshold. The method may further include segmenting, based on the segmentation threshold, the image.

Abstract: The present disclosure provides a method for automated image acquisition and imaging processing. The method may include obtaining imaging information of an object. The method may include determining, based on the imaging information, at least target device positioning information of an imaging device. The method may include causing, based on the target device positioning information of the imaging device, the imaging device to be positioned to perform the image acquisition. The method may include providing, based on the inspection information, guidance information, the guidance information being configured to guide positioning of the object. The method may also include obtaining a target image from an imaging operation by the imaging device. Further, the method may include determining a target image processing algorithm of a medical image.

Abstract: A region-of-interest extraction method and an apparatus, and a device, a system and a storage medium are provided. The region-of-interest extraction method is applied to an X-ray imaging system. The X-ray imaging system includes a detector and an array X-ray source, the array X-ray source includes a plurality of X-ray sources having different projection angles, and imaging regions corresponding to the plurality of X-ray sources are formed on the detector.

Abstract: A method may include obtaining an original image. The method may also include determining a plurality of decomposition coefficients of the original image by decomposing the original image. The method may also include determining at least one enhancement coefficient by performing enhancement to at least one of the plurality of decomposition coefficients using a coefficient enhancement model. The method may also include generating an enhanced image corresponding to the original image based on the at least one enhancement coefficient.

Abstract: A method and a system for medical imaging may be provided. A first optical image of a target subject that includes a target region to be scanned or treated by a medical device may be obtained. At least one body part boundary and at least one feature point of the target subject may be identified using at least one target recognition model from the first optical image. An image region corresponding to the target region of the target subject may be identified from the first optical image based on the at least one body part boundary and the at least one feature point. At least one first edge of the image region may be determined based on the at least one body part boundary, and at least one second edge of the image region may be determined based on the at least one feature point.