Search

The development of high-resolution remote sensing satellites has provided great convenience for research work related to remote sensing. Segmentation and extraction of specific targets are essential tasks when facing the vast and complex remote sensing images. Recently, the introduction of Segment Anything Model (SAM) provides a universal pre-training model for image segmentation tasks. While the direct application of SAM to remote sensing image segmentation tasks does not yield satisfactory results, we propose RSAM-Seg, which stands for Remote Sensing SAM with Semantic Segmentation, as a tailored modification of SAM for the remote sensing field and eliminates the need for manual intervention to provide prompts. Adapter-Scale, a set of supplementary scaling modules, are proposed in the multi-head attention blocks of the encoder part of SAM. Furthermore, Adapter-Feature are inserted between the Vision Transformer (ViT) blocks. These modules aim to incorporate high-frequency image information and image embedding features to generate image-informed prompts. Experiments are conducted on four distinct remote sensing scenarios, encompassing cloud detection, field monitoring, building detection and road mapping tasks . The experimental results not only showcase the improvement over the original SAM and U-Net across cloud, buildings, fields and roads scenarios, but also highlight the capacity of RSAM-Seg to discern absent areas within the ground truth of certain datasets, affirming its potential as an auxiliary annotation method. In addition, the performance in few-shot scenarios is commendable, underscores its potential in dealing with limited datasets., Comment: 12 pages, 11 figures

Node Importance Estimation (NIE) is a task of inferring importance scores of the nodes in a graph. Due to the availability of richer data and knowledge, recent research interests of NIE have been dedicating to knowledge graphs for predicting future or missing node importance scores. Existing state-of-the-art NIE methods train the model by available labels, and they consider every interested node equally before training. However, the nodes with higher importance often require or receive more attention in real-world scenarios, e.g., people may care more about the movies or webpages with higher importance. To this end, we introduce Label Informed ContrAstive Pretraining (LICAP) to the NIE problem for being better aware of the nodes with high importance scores. Specifically, LICAP is a novel type of contrastive learning framework that aims to fully utilize the continuous labels to generate contrastive samples for pretraining embeddings. Considering the NIE problem, LICAP adopts a novel sampling strategy called top nodes preferred hierarchical sampling to first group all interested nodes into a top bin and a non-top bin based on node importance scores, and then divide the nodes within top bin into several finer bins also based on the scores. The contrastive samples are generated from those bins, and are then used to pretrain node embeddings of knowledge graphs via a newly proposed Predicate-aware Graph Attention Networks (PreGAT), so as to better separate the top nodes from non-top nodes, and distinguish the top nodes within top bin by keeping the relative order among finer bins. Extensive experiments demonstrate that the LICAP pretrained embeddings can further boost the performance of existing NIE methods and achieve the new state-of-the-art performance regarding both regression and ranking metrics. The source code for reproducibility is available at https://github.com/zhangtia16/LICAP, Comment: Accepted by IEEE TNNLS

In autonomous driving, the hybrid strategy of deep reinforcement learning and cooperative adaptive cruise control (CACC) can fully utilize the advantages of the two algorithms and significantly improve the performance of car following. However, it is challenging for the traditional hybrid strategy based on fixed coefficients to adapt to mixed traffic flow scenarios, which may decrease the performance and even lead to accidents. To address the above problems, a hybrid car following strategy based on an adaptive Kalman Filter is proposed by regarding CACC and Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithms. Different from traditional hybrid strategy based on fixed coefficients, the Kalman gain H, using as an adaptive coefficient, is derived from multi-timestep predictions and Monte Carlo Tree Search. At the end of study, simulation results with 4157745 timesteps indicate that, compared with the TD3 and HCFS algorithms, the proposed algorithm in this study can substantially enhance the safety of car following in mixed traffic flow without compromising the comfort and efficiency., Comment: 32pages,13figures

The bumblebee model is an extension of the Einstein-Maxwell theory that allows for the spontaneous breaking of the Lorentz symmetry of the spacetime. In this paper, we study the quasinormal modes of the spherical black holes in this model that are characterized by a global monopole. We analyze the two cases with a vanishing cosmological constant or a negative one (the anti-de Sitter case). We find that the black holes are stable under the perturbation of a massless scalar field. However, both the Lorentz symmetry breaking and the global monopole have notable impacts on the evolution of the perturbation. The Lorentz symmetry breaking may prolong or shorten the decay of the perturbation according to the sign of the breaking parameter. The global monopole, on the other hand, has different effects depending on whether a nonzero cosmological constant presences: it reduces the damping of the perturbations for the case with a vanishing cosmological constant, but has little influence for the anti-de Sitter case., Comment: accepted by Eur. Phys. J. C

Training models with varying capacities can be advantageous for deploying them in different scenarios. While high-capacity models offer better performance, low-capacity models require fewer computing resources for training and inference. In this work, we propose a novel one-stop training framework to jointly train high-capacity and low-capactiy models. This framework consists of two composite model architectures and a joint training algorithm called Two-Stage Joint-Training (TSJT). Unlike knowledge distillation, where multiple capacity models are trained from scratch separately, our approach integrates supervisions from different capacity models simultaneously, leading to faster and more efficient convergence. Extensive experiments on the multilingual machine translation benchmark WMT10 show that our method outperforms low-capacity baseline models and achieves comparable or better performance on high-capacity models. Notably, the analysis demonstrates that our method significantly influences the initial training process, leading to more efficient convergence and superior solutions.

The paper investigates the features of emissions and fuel consumption (EFC) in a car-following (CF) platoon based on two experimental datasets. Four classical EFC models are employed and a universal concave growth pattern of the EFC along a platoon has been demonstrated. A general framework of coupling EFC and CF models is tested by calibrating and simulating three classical CF models. This work first demonstrates that, at vehicle-pair level, all models perform well on EFC prediction. The intelligent driver model outperforms the other CF models on calibration accuracy, but this is not true on EFC prediction. Second, at platoon level, the predicted EFC is nearly constant along the platoon which qualitatively differs from the experimental observation. The investigation highlights that accurate estimations at vehicle level may be insufficient for analysis at platoon level due to the significant role of oscillation growth and evolution in EFC estimation.

Knowledge graphs (KGs) have received increasing attention due to its wide applications on natural language processing. However, its use case on temporal question answering (QA) has not been well-explored. Most of existing methods are developed based on pre-trained language models, which might not be capable to learn \emph{temporal-specific} presentations of entities in terms of temporal KGQA task. To alleviate this problem, we propose a novel \textbf{T}ime-aware \textbf{M}ultiway \textbf{A}daptive (\textbf{TMA}) fusion network. Inspired by the step-by-step reasoning behavior of humans. For each given question, TMA first extracts the relevant concepts from the KG, and then feeds them into a multiway adaptive module to produce a \emph{temporal-specific} representation of the question. This representation can be incorporated with the pre-trained KG embedding to generate the final prediction. Empirical results verify that the proposed model achieves better performance than the state-of-the-art models in the benchmark dataset. Notably, the Hits@1 and Hits@10 results of TMA on the CronQuestions dataset's complex questions are absolutely improved by 24\% and 10\% compared to the best-performing baseline. Furthermore, we also show that TMA employing an adaptive fusion mechanism can provide interpretability by analyzing the proportion of information in question representations., Comment: ICASSP 2023

Recent experimental and empirical observations have demonstrated that stochasticity plays a critical role in car following (CF) dynamics. To reproduce the observations, quite a few stochastic CF models have been proposed. However, while calibrating the deterministic CF models is well investigated, studies on how to calibrate the stochastic models are lacking. Motivated by this fact, this paper aims to address this fundamental research gap. Firstly, the CF experiment under the same driving environment is conducted and analyzed. Based on the experimental results, we test two previous calibration methods, i.e., the method to minimize the Multiple Runs Mean (MRMean) error and the method of maximum likelihood estimation (MLE). Deficiencies of the two methods have been identified. Next, we propose a new method to minimize the Multiple Runs Minimum (MRMin) error. Calibration based on the experimental data and the synthetic data demonstrates that the new method outperforms the two previous methods. Furthermore, the mechanisms of different methods are explored from the perspective of error analysis. The analysis indicates that the new method can be regarded as a nested optimization model. The method separates the aleatoric errors caused by stochasticity from the epistemic error caused by parameters, and it is able to deal with the two kinds of errors effectively. Finally, we find that under the calibration framework of stochastic CF models, the calibrated parameter set using spacing as MoP may not always outperform that using velocity as MoP. These findings are expected to enhance the understanding of the role of stochasticity in CF dynamics where the new calibration framework for stochastic CF models is established., Comment: 34 pages, 13 figures, 17 tables

Accurate modeling of lower-level controller plays an important role in the traffic flow of automated vehicles (AVs). However, there lacks enough attention with this respect. To address this issue, we conduct a field experiment with two vehicles that are equipped with developable autonomous driving system, where one can customize the upper-level control algorithm. Based on the field experimental data, a new lower-level control model is developed and compared with two widely used ones. The comparison results show that the proposed model outperforms the two previous models in capturing the observed actual acceleration, especially the troughs of the acceleration time series. Furthermore, theoretical analysis indicates that comparing with the proposed model, the two previous models significantly overestimate the stability region of the traffic flow of the AVs and the capacity of stable traffic flow. Our study is expected to further shed light on the importance of accurate lower-level control modeling.

Even though the large-scale language models have achieved excellent performances, they suffer from various adversarial attacks. A large body of defense methods has been proposed. However, they are still limited due to redundant attack search spaces and the inability to defend against various types of attacks. In this work, we present a novel fine-tuning approach called \textbf{RO}bust \textbf{SE}letive fine-tuning (\textbf{ROSE}) to address this issue. ROSE conducts selective updates when adapting pre-trained models to downstream tasks, filtering out invaluable and unrobust updates of parameters. Specifically, we propose two strategies: the first-order and second-order ROSE for selecting target robust parameters. The experimental results show that ROSE achieves significant improvements in adversarial robustness on various downstream NLP tasks, and the ensemble method even surpasses both variants above. Furthermore, ROSE can be easily incorporated into existing fine-tuning methods to improve their adversarial robustness further. The empirical analysis confirms that ROSE eliminates unrobust spurious updates during fine-tuning, leading to solutions corresponding to flatter and wider optima than the conventional method. Code is available at \url{https://github.com/jiangllan/ROSE}., Comment: Accepted to EMNLP 2022. Code is available at https://github.com/jiangllan/ROSE

Anchor-free detectors basically formulate object detection as dense classification and regression. For popular anchor-free detectors, it is common to introduce an individual prediction branch to estimate the quality of localization. The following inconsistencies are observed when we delve into the practices of classification and quality estimation. Firstly, for some adjacent samples which are assigned completely different labels, the trained model would produce similar classification scores. This violates the training objective and leads to performance degradation. Secondly, it is found that detected bounding boxes with higher confidences contrarily have smaller overlaps with the corresponding ground-truth. Accurately localized bounding boxes would be suppressed by less accurate ones in the Non-Maximum Suppression (NMS) procedure. To address the inconsistency problems, the Dynamic Smooth Label Assignment (DSLA) method is proposed. Based on the concept of centerness originally developed in FCOS, a smooth assignment strategy is proposed. The label is smoothed to a continuous value in [0, 1] to make a steady transition between positive and negative samples. Intersection-of-Union (IoU) is predicted dynamically during training and is coupled with the smoothed label. The dynamic smooth label is assigned to supervise the classification branch. Under such supervision, quality estimation branch is naturally merged into the classification branch, which simplifies the architecture of anchor-free detector. Comprehensive experiments are conducted on the MS COCO benchmark. It is demonstrated that, DSLA can significantly boost the detection accuracy by alleviating the above inconsistencies for anchor-free detectors. Our codes are released at https://github.com/YonghaoHe/DSLA., Comment: single column, 33 pages, 7 figures, accepted by Pattern Recognition