Hardware prefetching is one of the most widely-used techniques for hiding long data access latency. To address the challenges faced by hardware prefetching, architects have proposed to detect and exploit the spatial locality at the granularity of spatial region. When a new region is activated, they try to find similar previously accessed regions for footprint prediction based on system-level environmental features such as the trigger instruction or data address. However, we find that such context-based prediction cannot capture the essential characteristics of access patterns, leading to limited flexibility, practicality and suboptimal prefetching performance. In this paper, inspired by the temporal property of memory accessing, we note that the temporal correlation exhibited within the spatial footprint is a key feature of spatial patterns. To this end, we propose Gaze, a simple and efficient hardware spatial prefetcher that skillfully utilizes footprint-internal temporal correlations to efficiently characterize spatial patterns. Meanwhile, we observe a unique unresolved challenge in utilizing spatial footprints generated by spatial streaming, which exhibit extremely high access density. Therefore, we further enhance Gaze with a dedicated two-stage approach that mitigates the over-prefetching problem commonly encountered in conventional schemes. Our comprehensive and diverse set of experiments show that Gaze can effectively enhance the performance across a wider range of scenarios. Specifically, Gaze improves performance by $\mathbf{5. 7 \%}$ and 5.4% at single-core, 11.4% and $\mathbf{8. 8 \%}$ at eight-core, compared to most recent low-cost solutions PMP and vBerti.

In robotic teleoperation systems, the effectiveness of teleoperation relies heavily on the operator performing control actions, underscoring the importance of master device design. Breaking away from the mechanical framework that relies on complex structures and restricts workspace, this paper adopts a new magnetic method and introduces a wireless magnetic master device (MMD) with an expandable, singularity-free workspace, designed to enhance maneuverability and operational reach. Specifically, the concept and development of the MMD for robotic teleoperation are presented, followed by an exploration of key technical aspects, including a six-degree-of-freedom (DoF) tracking approach using dual magnetic localization and controller design for the robotic arm using the MMD. Finally, the proposed method’s feasibility and accuracy are validated through magnetic localization and teleoperation control tests. Results indicate that the MMD achieves six-DoF localization with an average Euclidean localization error ofet= 0.0018 m and an average Geodesic distance error ofeR= 0.0604 rad within the central region of the workspace, and six-DoF manipulation (velocityet= 0.0062 m andeR=0.2091 rad in tests. Furthermore, the overall teleoperation control framework achieves tracking accuracy withet= 0.0085 m andeR= 0.1475 rad in robotic calligraphy task; achieves a broad scanning range of 0.2746 m along they-axis in robotic ultrasound scanning task; achieves a large-angle rotation of 9.15 rad in robotic clamp rotating task. Overall, this research advances master device design, aiming to inspire further innovation in master-follower teleoperation control for human-machine systems.

Content Delivery Network (CDN) has been widely used to provide data transmission services to end users. The edge cache servers are important components in CDN, and their hit ratios significantly influence the quality of cache service. However, edge caches are shared by multiple tenants (i.e., Internet Content Providers or ICPs) and the resource contention among tenants presents a huge challenge to improve the cache performance. Cache partitioning is a common method to deal with this chal-lenge, and several approaches have been proposed but still have some drawbacks. Existing methods bring non-negligible temporal and spatial overheads while obtaining features. Although some learning based methods have reduced these costs, the learning model convergence is slow due to the large searching space. To address the above problems, we propose a lightweight Reinforcement Learning based Cache Partitioning Approach (RL-Cache), which increases overall hit ratios of edge cache servers in CDN. The core of RL-Cache is a new feature named Compulsory Miss Ratio (CMR). It can be obtained in linear complexity and reflect the tenants' demand of cache space. To demonstrate the effectiveness of our approach, we not only utilize open-source traces from industrial CDNs but also collect real-world workloads from Tencent Cloud CDN. We develop a simulator to conduct several experiments driven by various traces. The experimental results show that compared to the commonly used methods, RL-Cache reduces the upstream traffic by 12.6% on average and improves the hit ratio by up to 4%.

This article presents a control strategy to carry out multiobject manipulation on a novelsoft robotictable (SoTa), which is a new form of the planar distributed manipulator. Manipulating multiple delicate objects simultaneously is an attractive feature of SoTa. The challenge here is to coordinate multiple objects in a confined planar space while avoiding interference with each other. The SoTa system adopts a manipulation strategy that includes a planning and a tracking stage for the purpose of sorting objects. The planning stage consists of two phases: 1) discrete path planning to find a path for each object on a grid map with respect to time; 2) trajectory generation to optimize and produce workable trajectories for SoTa. In the discrete path planning phase, a hierarchical searching method based on the time-varying potential field is proposed. Constraints of the SoTa system are modeled and incorporated into the path searching process. In the trajectory generation phase, a piecewise B-spline method is adopted to generate trajectories based on previously found discrete paths. Next, in the tracking stage, the objects are led to their goals along the trajectories ensuring safety and SoTa's capability. The performances of the proposed algorithm were simulated, analyzed, and compared with the conflict-based search method, which is optimal for multiagent path finding. A multiobject manipulation experiment of three objects on a$4\times 4$grid was conducted on the SoTa. The results demonstrated the effectiveness of the proposed control strategy in executing multiobject manipulations for sorting tasks on the SoTa.

In the recent storage market, Flash-based Solid State Drives (SSDs) have become high-performance alternatives to Hard Disk Drives (HDDs), dramatically increasing SSD shipments. To guarantee product reliability and quality to remain competitive, SSD manufacturers pay significant efforts in technology qualification and reliability design, especially in Manufacturing Self-Test (MST) processes. However, the cost of the MST process becomes more prominent as the memory density of SSD increases. In this paper, we study the MST data in over 20,000 SSDs and propose a novel and economical approach to dynamically reduce the MST overhead by proactive infant defect prediction based on Generative Adversarial Network-Attention based Spatial-Temporal Sequence-to-Sequence network (GAN-ASTSeq). It reduces the temporal cost by 80.2% (i.e., improves the efficiency by 4×) while maintaining an outstanding detection rate of defects.

Zero-shot stance detection (ZSSD) is challenging as it requires detecting the stance of previously unseen targets during the inference stage. Being able to detect the target-related transferable stance features from the training data is arguably an important step in ZSSD. Generally speaking, stance features can be grouped into target-invariant and target-specific categories. Target-invariant stance features carry the same stance regardless of the targets they are associated with. On the contrary, target-specific stance features only co-occur with certain targets. As such, it is important to distinguish these two types of stance features when learning stance features of unseen targets. To this end, in this paper, we revisit ZSSD from a novel perspective by developing an effective approach to distinguish the types (target-invariant/-specific) of stance features, so as to better learn transferable stance features. To be specific, inspired by self-supervised learning, we frame the stance-feature-type identification as a pretext task in ZSSD. Furthermore, we devise a novel hierarchical contrastive learning strategy to capture the correlation and difference between target-invariant and -specific features and further among different stance labels. This essentially allows the model to exploit transferable stance features more effectively for representing the stance of previously unseen targets. Extensive experiments on three benchmark datasets show that the proposed framework achieves the state-of-the-art performance in ZSSD.