Haozhi Li

Haozhi Li, Tariq Elahi

High latency is a critical limitation within the Tor network that has a negative impact on web application responsiveness. A key factor exacerbating Tor latency is the creation of lengthy circuits that span across geographically distant regions, causing significant transmission delays. A common solution involves modifying Tor's circuit-building process to reduce the likelihood of selecting lengthy circuits. However, this strategy compromises Tor's routing randomness, increasing the risk of deanonymization. Reducing Tor's latency while minimizing security degradation presents a challenge. This paper proposes and investigates SaTor, a satellite-assisted routing scheme to reduce Tor latency. By equipping Tor relays with satellite network access, SaTor could accelerate slow circuits via satellite transmission, without biasing the existing path selection process. Our performance evaluation, using a simulator we developed along with real-world measurements, shows that over the long term, SaTor provides an expected speed-up of 21.8 ms for over 40% of circuits, with only 100 top relays equipped with satellite service. Our research uncovers a viable way to overcome Tor's latency bottleneck, serving as a practical reference for its future enhancement.

Longyao Liu, Bo Ma, Yulin Zhang et al.

Few-shot object detection (FSOD) aims at learning a detector that can fast adapt to previously unseen objects with scarce annotated examples, which is challenging and demanding. Existing methods solve this problem by performing subtasks of classification and localization utilizing a shared component (e.g., RoI head) in the detector, yet few of them take the distinct preferences of two subtasks towards feature embedding into consideration. In this paper, we carefully analyze the characteristics of FSOD, and present that a general few-shot detector should consider the explicit decomposition of two subtasks, as well as leveraging information from both of them to enhance feature representations. To the end, we propose a simple yet effective Adaptive Fully-Dual Network (AFD-Net). Specifically, we extend Faster R-CNN by introducing Dual Query Encoder and Dual Attention Generator for separate feature extraction, and Dual Aggregator for separate model reweighting. Spontaneously, separate state estimation is achieved by the R-CNN detector. Besides, for the acquisition of enhanced feature representations, we further introduce Adaptive Fusion Mechanism to adaptively perform feature fusion in different subtasks. Extensive experiments on PASCAL VOC and MS COCO in various settings show that, our method achieves new state-of-the-art performance by a large margin, demonstrating its effectiveness and generalization ability.