This work focuses on low-altitude networks in the field of wireless communications. To ensure ground coverage performance of the existing ground base station, a certain downtilt angle is set for an antenna, and a back beam of the antenna points to a space domain. According to the current low-altitude network spatial domain test, the low-altitude spatial domain has a large number of radio signals, and the overlapping coverage is high. Under the same Reference Signal Received Power (RSRP), the Signal to Interference plus Noise Ratio (SINR) value that represents the radio signal quality is lower than that on the ground. In addition, a service requirement of a low-altitude network focuses on an uplink high-bandwidth service such as video surveillance and security rescue. However, an existing low-altitude communication network cannot meet a service requirement. This paper proposes a coverage enhancement method for a low-altitude network oriented to a large uplink bandwidth, and aims at the problems of low-altitude spatial domain signal clutter and high uplink background noise, and abandons the existing solution of adding a power amplifier or installing a large gain antenna for improving uplink coverage performance. The omnidirectional antenna carried by the Unmanned Aerial Vehicle (UAV) is replaced with a directional antenna, and the multi-feed lens antenna is used to transmit and receive RF signals in all directions. This increases the uplink antenna gain and improves the uplink coverage performance. In addition, the uplink achieves a higher Modulation and Coding Scheme (MCS). Multi-feed intelligent transmit control can also be used to increase the uplink 2T transmit ratio and finally increase the uplink transmission rate to meet the large uplink bandwidth requirements of video backhaul services.

In this paper, we introduce UI-Genie, a self-improving framework addressing two key challenges in GUI agents: verification of trajectory outcome is challenging and high-quality training data are not scalable. These challenges are addressed by a reward model and a self-improving pipeline, respectively. The reward model, UI-Genie-RM, features an image-text interleaved architecture that efficiently processes historical context and unifies action-level and task-level rewards. To support the training of UI-Genie-RM, we develop deliberately-designed data generation strategies including rule-based verification, controlled trajectory corruption, and hard negative mining. To address the second challenge, a self-improvement pipeline progressively expands solvable complex GUI tasks by enhancing both the agent and reward models through reward-guided exploration and outcome verification in dynamic environments. For training the model, we generate UI-Genie-RM-517k and UI-Genie-Agent-16k, establishing the first reward-specific dataset for GUI agents while demonstrating high-quality synthetic trajectory generation without manual annotation. Experimental results show that UI-Genie achieves state-of-the-art performance across multiple GUI agent benchmarks with three generations of data-model self-improvement. We open-source our complete framework implementation and generated datasets to facilitate further research in https://github.com/Euphoria16/UI-Genie.

Document image dewarping is a crucial task in computer vision with numerous practical applications. The control point method, as a popular image dewarping approach, has attracted attention due to its simplicity and efficiency. However, inaccurate control point prediction due to varying background noises and deformation types can result in unsatisfactory performance. To address these issues, we propose a robust document dewarping approach for real-life images, namely DocReal, which utilizes Enet to effectively remove background noise and an attention-enhanced control point (AECP) module to better capture local deformations. Moreover, we augment the training data by synthesizing 2D images with 3D deformations and additional deformation types. Our proposed method achieves state-of-the-art performance on the DocUNet benchmark and a newly proposed benchmark of 200 Chinese distorted images, exhibiting superior dewarping accuracy, OCR performance, and robustness to various types of image distortion.