Search

Anderson localization and non-Hermitian skin effect are two paradigmatic wave localization phenomena, resulting from wave interference and the intrinsic non-Hermitian point gap, respectively. In this study, we unveil a novel localization phenomenon associated with long-range asymmetric coupling, termed scale-tailored localization, where the number of induced localized modes and their localization lengths scale exclusively with the coupling range. We show that the long-range coupling fundamentally reshapes the energy spectra and eigenstates by creating multiple connected paths on the lattice. Furthermore, we present experimental observations of scale-tailored localization in non-Hermitian electrical circuits utilizing adjustable voltage followers and switches. The circuit admittance spectra possess separate point-shaped and loop-shaped components in the complex energy plane, corresponding respectively to skin modes and scale-tailored localized states. Our findings not only expand and deepen the understanding of peculiar effects induced by non-Hermiticity but also offer a feasible experimental platform for exploring and controlling wave localizations., Comment: 10+16 pages, 5+11 figures

Lane detection has made significant progress in recent years, but there is not a unified architecture for its two sub-tasks: 2D lane detection and 3D lane detection. To fill this gap, we introduce B\'{e}zierFormer, a unified 2D and 3D lane detection architecture based on B\'{e}zier curve lane representation. B\'{e}zierFormer formulate queries as B\'{e}zier control points and incorporate a novel B\'{e}zier curve attention mechanism. This attention mechanism enables comprehensive and accurate feature extraction for slender lane curves via sampling and fusing multiple reference points on each curve. In addition, we propose a novel Chamfer IoU-based loss which is more suitable for the B\'{e}zier control points regression. The state-of-the-art performance of B\'{e}zierFormer on widely-used 2D and 3D lane detection benchmarks verifies its effectiveness and suggests the worthiness of further exploration., Comment: ICME 2024, 11 pages, 8 figures

The conventional circuit diagrams and graph-based circuit theory are used for the phase-independent circuits such as resistor-inductor-capacitor (RLC) circuits and semiconductor transistor circuits, rather than for the phase-dependent circuits such as Josephson junction circuits and quantum-phase-slip (QPS) junction circuits. To unify the analyses for both phase-independent and phase-dependent electric circuits, we develop an electromagnetic-field-based circuit theory and invent the electric-charge-flow (ECF) and the magnetic-flux-flow (MFF) diagrams to derive the general circuit equations and visualizes the dynamics of circuit devices for all kinds of electric circuits. ECF and MFF diagrams enable electric circuits to be described like the molecules composed of two kinds of atoms; they are new languages for circuit design and analysis, promising for machine-learning., Comment: 41 pages, 35 figures

Flux transformers and superconducting quantum interference devices (SQUIDs) are two key superconducting devices used in the cutting-edge magnetic-field sensor systems, such as magnetocardiography (MCG) and magnetoencephalography (MEG). They are superconductor integrated circuits enabled by two superconductor elements, superconductor wire and Josephson junction, based on Meissner effect and Josephson effect. The principles of SQUID sensors are unfamiliar for electronics engineers who are trained with semiconductor electronics, due to the superconductor physics. We present an introduction to SQUID sensors, ranged from superconductor elements to the superconductor-semiconductor hybrid system. It is demonstrated that, SQUIDs inside are the resistor-inductor-capacitor (RLC) network driven by Josephson currents, and share the common network equation with normal RLC circuits; SQUID outside are magnetic-field-effect transistors (MFETs), and are dual to the semiconductor FETs in signal conversions; SQUID magnetic-field sensors are linear current amplifiers, as semiconductor FET electric-field sensors are linear voltage amplifiers., Comment: 53 pages, 50 figures, Principles covered from Josephson junction to SQUID sensor system

Recently, Transformer-based text detection techniques have sought to predict polygons by encoding the coordinates of individual boundary vertices using distinct query features. However, this approach incurs a significant memory overhead and struggles to effectively capture the intricate relationships between vertices belonging to the same instance. Consequently, irregular text layouts often lead to the prediction of outlined vertices, diminishing the quality of results. To address these challenges, we present an innovative approach rooted in Sparse R-CNN: a cascade decoding pipeline for polygon prediction. Our method ensures precision by iteratively refining polygon predictions, considering both the scale and location of preceding results. Leveraging this stabilized regression pipeline, even employing just a single feature vector to guide polygon instance regression yields promising detection results. Simultaneously, the leverage of instance-level feature proposal substantially enhances memory efficiency (>50% less vs. the state-of-the-art method DPText-DETR) and reduces inference speed (>40% less vs. DPText-DETR) with minor performance drop on benchmarks.

Superconducting quantum interference devices (SQUIDs), single flux-quantum (SFQ) logic circuits, and quantum Josephson junction circuits have been developed into a family of superconductor integrated circuit, and are widely applied for subtle magnetic-field measurements, energy-efficient computing, and quantum computing, respectively. They are Josephson junction networks composed of Josephson junctions and normal resistor-inductor-capacitor (RLC) components, working with the fluxoid-quantization principle and Josephson effects to achieve unique flux-modulated dynamics and characteristics; they react to the vector potential of magnetic fields rather than the electric potential. However, the conventional circuit diagrams and nodal analysis methods focus on the electric charges flowing though branches and nodes, ignoring dynamics of the magnetic fluxes flowing from loop to loop. This article introduces a general flux-based circuit theory to unify the analyses of Josephson junction circuits and normal RLC circuits. This theory presents a magnetic-flux-generator (MFG) concept to unify Josephson junctions and normal circuit elements, and abstract both Josephson junction circuits and normal RLC circuits as MFG network; it derives a general network equation to describe dynamics of Josephson junction circuits, and invents a kind of magnetic-flux flow (MFF) diagram to depict the working principles of magnetic-flux flows inside Josephson junction circuits. The flux-based theory is complementary to the conventional circuit theories in the design and analysis of superconductor integrated circuits., Comment: 35 pages, 52 figures

Vision-centric Bird's-Eye View (BEV) representation is essential for autonomous driving systems (ADS). Multi-frame temporal fusion which leverages historical information has been demonstrated to provide more comprehensive perception results. While most research focuses on ego-centric maps of fixed settings, long-range local map generation remains less explored. This work outlines a new paradigm, named NeMO, for generating local maps through the utilization of a readable and writable big map, a learning-based fusion module, and an interaction mechanism between the two. With an assumption that the feature distribution of all BEV grids follows an identical pattern, we adopt a shared-weight neural network for all grids to update the big map. This paradigm supports the fusion of longer time series and the generation of long-range BEV local maps. Furthermore, we release BDD-Map, a BDD100K-based dataset incorporating map element annotations, including lane lines, boundaries, and pedestrian crossing. Experiments on the NuScenes and BDD-Map datasets demonstrate that NeMO outperforms state-of-the-art map segmentation methods. We also provide a new scene-level BEV map evaluation setting along with the corresponding baseline for a more comprehensive comparison.

Review summarization is a non-trivial task that aims to summarize the main idea of the product review in the E-commerce website. Different from the document summary which only needs to focus on the main facts described in the document, review summarization should not only summarize the main aspects mentioned in the review but also reflect the personal style of the review author. Although existing review summarization methods have incorporated the historical reviews of both customer and product, they usually simply concatenate and indiscriminately model this two heterogeneous information into a long sequence. Moreover, the rating information can also provide a high-level abstraction of customer preference, it has not been used by the majority of methods. In this paper, we propose the Heterogeneous Historical Review aware Review Summarization Model (HHRRS) which separately models the two types of historical reviews with the rating information by a graph reasoning module with a contrastive loss. We employ a multi-task framework that conducts the review sentiment classification and summarization jointly. Extensive experiments on four benchmark datasets demonstrate the superiority of HHRRS on both tasks.

GNSS and LiDAR odometry are complementary as they provide absolute and relative positioning, respectively. Their integration in a loosely-coupled manner is straightforward but is challenged in urban canyons due to the GNSS signal reflections. Recent proposed 3D LiDAR-aided (3DLA) GNSS methods employ the point cloud map to identify the non-line-of-sight (NLOS) reception of GNSS signals. This facilitates the GNSS receiver to obtain improved urban positioning but not achieve a sub-meter level. GNSS real-time kinematics (RTK) uses carrier phase measurements to obtain decimeter-level positioning. In urban areas, the GNSS RTK is not only challenged by multipath and NLOS-affected measurement but also suffers from signal blockage by the building. The latter will impose a challenge in solving the ambiguity within the carrier phase measurements. In the other words, the model observability of the ambiguity resolution (AR) is greatly decreased. This paper proposes to generate virtual satellite (VS) measurements using the selected LiDAR landmarks from the accumulated 3D point cloud maps (PCM). These LiDAR-PCM-made VS measurements are tightly-coupled with GNSS pseudorange and carrier phase measurements. Thus, the VS measurements can provide complementary constraints, meaning providing low-elevation-angle measurements in the across-street directions. The implementation is done using factor graph optimization to solve an accurate float solution of the ambiguity before it is fed into LAMBDA. The effectiveness of the proposed method has been validated by the evaluation conducted on our recently open-sourced challenging dataset, UrbanNav. The result shows the fix rate of the proposed 3DLA GNSS RTK is about 30% while the conventional GNSS-RTK only achieves about 14%. In addition, the proposed method achieves sub-meter positioning accuracy in most of the data collected in challenging urban areas.

The non-Hermitian systems with the non-Hermitian skin effect (NHSE) are very sensitive to the imposed boundary conditions and lattice size, which leads to size-dependent non-Hermitian skin effects. Here, we report the experimental observation of NHSE with different boundary conditions and different lattice size in a unidirectional hopping model based on a circuit platform. The circuit admittance spectra and corresponding eigenstates are very sensitive to the presence of the boundary. Meanwhile, our experimental results show how the lattice size and boundary terms together affect the strength of NHSE. Therefore, our electric circuit provides a good platform to observe size-dependent boundary effects in non-Hermitian systems., Comment: 8 pages, 4 figures, 4 tables

Josephson junction circuits, such as superconducting quantum interference devices (SQUIDs) and single-flux-quantum (SFQ) circuits, have been applied in both analog and digital systems for their ultralow-noise, high-speed, and power-efficient features. However, their analyses are not well supported by the conventional circuit diagrams and the charge-based analysis methods, due to the Josephson effects and flux quantization phenomena. This article introduces a type of magnetic-flux-flow (MFF) diagrams for the design and analysis of Josephson junction circuits. The MFF diagram of a Josephson junction circuit is the graphical expression of circuit equations. It not only gives the behaviors of Josephson junctions and superconducting loops new physical meanings, but also visualizes how the flux quanta are transferred among superconducting loops. The advantages of MFF diagrams are demonstrated in the design and analysis of typical Josephson junction circuits. It is shown that the MFF diagram and flux-based analysis method are the indispensable complement to the conventional circuit methodologies for Josephson junction circuits., Comment: 7 pages, 19 figures

Reaching tasks with random targets and obstacles is a challenging task for robotic manipulators. In this study, we propose a novel model-free reinforcement learning approach based on proximal policy optimization (PPO) for training a deep policy to map the task space to the joint space of a 6-DoF manipulator. To facilitate the training process in a large workspace, we develop an efficient representation of environmental inputs and outputs. The calculation of the distance between obstacles and manipulator links is incorporated into the state representation using a geometry-based method. Additionally, to enhance the performance of the model in reaching tasks, we introduce the action ensembles method and design the policy to directly participate in value function updates in PPO. To overcome the challenges associated with training in real-robot environments, we develop a simulation environment in Gazebo to train the model as it produces a smaller Sim-to-Real gap compared to other simulators. However, training in Gazebo is time-intensive. To address this issue, we propose a Sim-to-Sim method to significantly reduce the training time. The trained model is then directly applied in a real-robot setup without fine-tuning. To evaluate the performance of the proposed approach, we perform several rounds of experiments in both simulated and real robots. We also compare the performance of the proposed approach with six baselines. The experimental results demonstrate the effectiveness of the proposed method in performing reaching tasks with and without obstacles. our method outperformed the selected baselines by a large margin in different reaching task scenarios. A video of these experiments has been attached to the paper as supplementary material.

The task of Human-Object Interaction~(HOI) detection could be divided into two core problems, i.e., human-object association and interaction understanding. In this paper, we reveal and address the disadvantages of the conventional query-driven HOI detectors from the two aspects. For the association, previous two-branch methods suffer from complex and costly post-matching, while single-branch methods ignore the features distinction in different tasks. We propose Guided-Embedding Network~(GEN) to attain a two-branch pipeline without post-matching. In GEN, we design an instance decoder to detect humans and objects with two independent query sets and a position Guided Embedding~(p-GE) to mark the human and object in the same position as a pair. Besides, we design an interaction decoder to classify interactions, where the interaction queries are made of instance Guided Embeddings (i-GE) generated from the outputs of each instance decoder layer. For the interaction understanding, previous methods suffer from long-tailed distribution and zero-shot discovery. This paper proposes a Visual-Linguistic Knowledge Transfer (VLKT) training strategy to enhance interaction understanding by transferring knowledge from a visual-linguistic pre-trained model CLIP. In specific, we extract text embeddings for all labels with CLIP to initialize the classifier and adopt a mimic loss to minimize the visual feature distance between GEN and CLIP. As a result, GEN-VLKT outperforms the state of the art by large margins on multiple datasets, e.g., +5.05 mAP on HICO-Det. The source codes are available at https://github.com/YueLiao/gen-vlkt., Comment: CVPR 2022

Context modeling plays a significant role in building multi-turn dialogue systems. In order to make full use of context information, systems can use Incomplete Utterance Rewriting(IUR) methods to simplify the multi-turn dialogue into single-turn by merging current utterance and context information into a self-contained utterance. However, previous approaches ignore the intent consistency between the original query and rewritten query. The detection of omitted or coreferred locations in the original query can be further improved. In this paper, we introduce contrastive learning and multi-task learning to jointly model the problem. Our method benefits from carefully designed self-supervised objectives, which act as auxiliary tasks to capture semantics at both sentence-level and token-level. The experiments show that our proposed model achieves state-of-the-art performance on several public datasets.

Building robust and general dialogue models for spoken conversations is challenging due to the gap in distributions of spoken and written data. This paper presents our approach to build generalized models for the Knowledge-grounded Task-oriented Dialogue Modeling on Spoken Conversations Challenge of DSTC-10. In order to mitigate the discrepancies between spoken and written text, we mainly employ extensive data augmentation strategies on written data, including artificial error injection and round-trip text-speech transformation. To train robust models for spoken conversations, we improve pre-trained language models, and apply ensemble algorithms for each sub-task. Typically, for the detection task, we fine-tune \roberta and ELECTRA, and run an error-fixing ensemble algorithm. For the selection task, we adopt a two-stage framework that consists of entity tracking and knowledge ranking, and propose a multi-task learning method to learn multi-level semantic information by domain classification and entity selection. For the generation task, we adopt a cross-validation data process to improve pre-trained generative language models, followed by a consensus decoding algorithm, which can add arbitrary features like relative \rouge metric, and tune associated feature weights toward \bleu directly. Our approach ranks third on the objective evaluation and second on the final official human evaluation.

Task-oriented dialogue systems have become overwhelmingly popular in recent researches. Dialogue understanding is widely used to comprehend users' intent, emotion and dialogue state in task-oriented dialogue systems. Most previous works on such discriminative tasks only models current query or historical conversations. Even if in some work the entire dialogue flow was modeled, it is not suitable for the real-world task-oriented conversations as the future contexts are not visible in such cases. In this paper, we propose to jointly model historical and future information through the posterior regularization method. More specifically, by modeling the current utterance and past contexts as prior, and the entire dialogue flow as posterior, we optimize the KL distance between these distributions to regularize our model during training. And only historical information is used for inference. Extensive experiments on two dialogue datasets validate the effectiveness of our proposed method, achieving superior results compared with all baseline models.

Robots often face situations where grasping a goal object is desirable but not feasible due to other present objects preventing the grasp action. We present a deep Reinforcement Learning approach to learn grasping and pushing policies for manipulating a goal object in highly cluttered environments to address this problem. In particular, a dual Reinforcement Learning model approach is proposed, which presents high resilience in handling complicated scenes, reaching an average of 98% task completion using primitive objects in a simulation environment. To evaluate the performance of the proposed approach, we performed two extensive sets of experiments in packed objects and a pile of object scenarios with a total of 1000 test runs in simulation. Experimental results showed that the proposed method worked very well in both scenarios and outperformed the recent state-of-the-art approaches. Demo video, trained models, and source code for the results reproducibility purpose are publicly available. https://sites.google.com/view/pushandgrasp/home, Comment: This paper has been accepted for publication at the ICRA2024 conference

Community Question Answering (CQA) is a well-defined task that can be used in many scenarios, such as E-Commerce and online user community for special interests. In these communities, users can post articles, give comment, raise a question and answer it. These data form the heterogeneous information sources where each information source have their own special structure and context (comments attached to an article or related question with answers). Most of the CQA methods only incorporate articles or Wikipedia to extract knowledge and answer the user's question. However, various types of information sources in the community are not fully explored by these CQA methods and these multiple information sources (MIS) can provide more related knowledge to user's questions. Thus, we propose a question-aware heterogeneous graph transformer to incorporate the MIS in the user community to automatically generate the answer. To evaluate our proposed method, we conduct the experiments on two datasets: $\text{MSM}^{\text{plus}}$ the modified version of benchmark dataset MS-MARCO and the AntQA dataset which is the first large-scale CQA dataset with four types of MIS. Extensive experiments on two datasets show that our model outperforms all the baselines in terms of all the metrics., Comment: Accepted by WSDM 2022; Code at https://github.com/gsh199449/HeteroQA

Emotion recognition in conversation (ERC) has attracted much attention in recent years for its necessity in widespread applications. Existing ERC methods mostly model the self and inter-speaker context separately, posing a major issue for lacking enough interaction between them. In this paper, we propose a novel Speaker and Position-Aware Graph neural network model for ERC (S+PAGE), which contains three stages to combine the benefits of both Transformer and relational graph convolution network (R-GCN) for better contextual modeling. Firstly, a two-stream conversational Transformer is presented to extract the coarse self and inter-speaker contextual features for each utterance. Then, a speaker and position-aware conversation graph is constructed, and we propose an enhanced R-GCN model, called PAG, to refine the coarse features guided by a relative positional encoding. Finally, both of the features from the former two stages are input into a conditional random field layer to model the emotion transfer.