Recent advances in few-shot novel-view synthesis based on 3D Gaussian Splatting (3DGS) have shown remarkable progress. Existing methods usually rely on carefully designed geometric regularizers to reinforce geometric supervision; however, applying multiple regularizers consistently across scenes is hard to tune and often degrades robustness. Consequently, generating reliable geometry from extremely sparse viewpoints remains a key challenge. To overcome this limitation, we introduce SREGS, a framework tailored for few-shot reconstruction whose contributions focus on two aspects: explicitly consistent geometry and multi-scale depth-guided optimization. Specifically, to explicitly optimize reconstruction consistency, we initialize the point cloud with 2D Gaussians, thereby enhancing depth consistency for the same Gaussian observed from different views. Secondly, we employ region-adaptive rapid densificationn to fill under-covered regions with additional representations, while an opacity-aware noise term injects stochasticity into each Gaussian to boost exploration in under-observed areas. In addition, to strengthen geometric refinement of the radiance field, we impose multi-scale depth constraints based on a monocular depth prior, performing geometric refinement from global to local scales and ensuring highly accurate reconstruction. Extensive experiments on LLFF, MipNeRF360, and Blender show that SREGS achieves higher synthesis quality with lower computational cost and demonstrates robust performance. The code is available at:https://github.com/LeeXiaoTong1/SREGS.

Large-scale deployment of Internet of Things (IoT) devices provides efficient data collection and control capabilities in the smart grid, while edge computing plays a key role in increasing the speed of data processing and reducing latency. The growth of edge computing in the smart grid is inevitably increasing the stability requirement in network transmission and the necessity of adequate network measurement to detect traffic anomalies. There are two challenges for measurement in IoT-enabled networks. The first challenge is the high processing speed and limited memory space. The second challenge is the varying network traffic. This paper studies how to use sketch techniques to fulfill the required capabilities of network measurement. Sketches have been considered as the most promising solution for network measurement in recent years, because they greatly optimize the speed and memory usage at the cost of small error. However, most sketches do not work well for varying network traffic. These sketches require to adjust their internal memory usage while the optimal point is sensitive to specified flow size distribution and memory size. To address this problem, we propose a machine learning empowered sketch framework. The proposed sketch framework trains a neural network model that online adjusts the memory usage based on a small sample of flow. We conduct data-driven simulations to evaluate the proposed framework. By comparing measurement results on public dataset, we can see that the proposed framework significantly improves the accuracy of measurement and anomaly detection without sacrificing the line-rate processing capability.

This work focuses on exploring the accuracy of heading estimation in x dead-reckoning (xDR) positioning methods across different time scales to address the limitations of pedestrian dead-reckoning (PDR) which relies solely on step analysis for location updates. To this end, we first give the relationship among position update frequency of PDR, sampling rate and step frequency in formula which shows the instability. And we propose a method utilizing an improved VIT-Encoder to mine the correlation features of Inertial Measurement Unit (IMU) sequences for heading estimation to eliminate the instability of PDR and evaluate their generalization performance. Results indicate that the model achieves accurate heading estimation on both training and test sets at larger time scales, with substantial disparities observed at smaller scales. Notably, the robustness of smaller-scale estimations mirrors or even surpasses that of longer time scales, revealing an intriguing finding. Additionally, the model exhibits high recognition accuracy across different time scales in straight trajectories, whereas shorter time scales demonstrate advantages at bends, offering higher fault tolerance for misestimations. Increasing the positioning update frequency to 10 Hz in xDR positioning methods is deemed feasible, which holds significant implications for enhancing the performance and user experience of indoor positioning systems, thereby promoting the development and application of indoor positioning technology.

Emotion-cause pair extraction is an emergent natural language processing task; the target is to extract all pairs of emotion clauses and corresponding cause clauses from unannotated emotion text. Previous studies have employed two-step approaches. However, this research may lead to error propagation across stages. In addition, previous studies did not correctly handle the situation where emotion clauses and cause clauses are the same clauses. To overcome these issues, the authors first use a multitask learning model that is based on graph from the perspective of sorting, which can simultaneously extract emotion clauses, cause clauses and emotion-cause pairs via an end-to-end strategy. Then the authors propose to convert text into graph structured data, and process this scenario through a unique graph convolutional neural network. Finally, the authors design a semantic decision mechanism to address the scenario in which there are multiple emotion-cause pairs in a text.

Capacity limit is a bottleneck for broader applications of blockchain systems. Scaling up capacity while preserving security and decentralization are major challenges in blockchain infrastructure design. In this paper, we design a proof of work-based mechanism by endowing directed acyclic graphs (DAG) with a novel structure so that peers can reach consensus at a large scale. At a high level, we break large blocks into smaller ones to improve utilization of broadcast network and embed a Nakamoto chain inside the DAG in a decent way to ensure security. We further exploit the DAG structure and design a mempool transaction assignment method. The method reduces the probability that a transaction is processed by multiple miners and hence improves processing efficiency. Without sacrificing security and decentralization, our design significant scales up capacity and also addresses important issues such as high latency and mining power concentration in existing blockchain systems.