arXiv subjects

Long Chen

Publications and source records attributed to Long Chen.

At least 451 records · Page 25Linked to original sources

arXiv2023

Structure, physical properties, and magnetically tunable topological phases in topological semimetal EuCuBi

A single material achieving multiple topological phases can provide potential application for topological spintronics, whereas the candidate materials are very limited. Here, we report the structure, physical properties, and possible emergence of multiple topological phases in the newly discovered, air-stable EuCuBi single crystal. EuCuBi crystallizes in a hexagonal space group P63/mmc (No. 194) in ZrBeSi-type structure with an antiferromagnetic (AFM) ground state below TN = 11.2 K. There is a competition between AFM and ferromagnetic (FM) interactions below TN revealed by electrical resistivity and magnetic susceptibility measurements. With the increasing magnetic field, EuCuBi evolves from the AFM ground state with a small amount of FM component, going through two possible metamagnetic phases, finally reaches the field-induced FM phase. Based on the first-principles calculations, we demonstrate that the Dirac, Weyl, and possible mirror Chern insulator can be achieved in EuCuBi by tuning the temperature and applying magnetic field, making EuCuBi a promising candidate for exploring multiple topological phases.

cond-mat.mtrl-sci↗

arXiv2023

Transformer Meets Boundary Value Inverse Problems

A Transformer-based deep direct sampling method is proposed for electrical impedance tomography, a well-known severely ill-posed nonlinear boundary value inverse problem. A real-time reconstruction is achieved by evaluating the learned inverse operator between carefully designed data and the reconstructed images. An effort is made to give a specific example to a fundamental question: whether and how one can benefit from the theoretical structure of a mathematical problem to develop task-oriented and structure-conforming deep neural networks? Specifically, inspired by direct sampling methods for inverse problems, the 1D boundary data in different frequencies are preprocessed by a partial differential equation-based feature map to yield 2D harmonic extensions as different input channels. Then, by introducing learnable non-local kernels, the direct sampling is recast to a modified attention mechanism. The new method achieves superior accuracy over its predecessors and contemporary operator learners and shows robustness to noises in benchmarks. This research shall strengthen the insights that, despite being invented for natural language processing tasks, the attention mechanism offers great flexibility to be modified in conformity with the a priori mathematical knowledge, which ultimately leads to the design of more physics-compatible neural architectures.

cs.LG↗

arXiv2023

Weakly-Supervised Temporal Article Grounding

Given a long untrimmed video and natural language queries, video grounding (VG) aims to temporally localize the semantically-aligned video segments. Almost all existing VG work holds two simple but unrealistic assumptions: 1) All query sentences can be grounded in the corresponding video. 2) All query sentences for the same video are always at the same semantic scale. Unfortunately, both assumptions make today's VG models fail to work in practice. For example, in real-world multimodal assets (eg, news articles), most of the sentences in the article can not be grounded in their affiliated videos, and they typically have rich hierarchical relations (ie, at different semantic scales). To this end, we propose a new challenging grounding task: Weakly-Supervised temporal Article Grounding (WSAG). Specifically, given an article and a relevant video, WSAG aims to localize all ``groundable'' sentences to the video, and these sentences are possibly at different semantic scales. Accordingly, we collect the first WSAG dataset to facilitate this task: YouwikiHow, which borrows the inherent multi-scale descriptions in wikiHow articles and plentiful YouTube videos. In addition, we propose a simple but effective method DualMIL for WSAG, which consists of a two-level MIL loss and a single-/cross- sentence constraint loss. These training objectives are carefully designed for these relaxed assumptions. Extensive ablations have verified the effectiveness of DualMIL.

cs.CV↗

arXiv2023

A gradient descent akin method for constrained optimization: algorithms and applications

We present a first-order method for solving constrained optimization problems. The method is derived from our previous work, a modified search direction method inspired by singular value decomposition. In this work, we simplify its computational framework to a ``gradient descent akin'' method (GDAM), i.e., the search direction is computed using a linear combination of the negative and normalized objective and constraint gradient. We give fundamental theoretical guarantees on the global convergence of the method. This work focuses on the algorithms and applications of GDAM. We present computational algorithms that adapt common strategies for the gradient descent method. We demonstrate the potential of the method using two engineering applications, shape optimization and sensor network localization. When practically implemented, GDAM is robust and very competitive in solving the considered large and challenging optimization problems.

math.OC↗

arXiv2023

Instance-wise or Class-wise? A Tale of Neighbor Shapley for Concept-based Explanation

Deep neural networks have demonstrated remarkable performance in many data-driven and prediction-oriented applications, and sometimes even perform better than humans. However, their most significant drawback is the lack of interpretability, which makes them less attractive in many real-world applications. When relating to the moral problem or the environmental factors that are uncertain such as crime judgment, financial analysis, and medical diagnosis, it is essential to mine the evidence for the model's prediction (interpret model knowledge) to convince humans. Thus, investigating how to interpret model knowledge is of paramount importance for both academic research and real applications.

cs.LG↗

arXiv2023

Transformed Primal-Dual Methods For Nonlinear Saddle Point Systems

A transformed primal-dual (TPD) flow is developed for a class of nonlinear smooth saddle point system. The flow for the dual variable contains a Schur complement which is strongly convex. Exponential stability of the saddle point is obtained by showing the strong Lyapunov property. Several TPD iterations are derived by implicit Euler, explicit Euler, implicit-explicit and Gauss-Seidel methods with accelerated overrelaxation of the TPD flow. Generalized to the symmetric TPD iterations, linear convergence rate is preserved for convex-concave saddle point systems under assumptions that the regularized functions are strongly convex. The effectiveness of augmented Lagrangian methods can be explained as a regularization of the non-strongly convexity and a preconditioning for the Schur complement. The algorithm and convergence analysis depends crucially on appropriate inner products of the spaces for the primal variable and dual variable. A clear convergence analysis with nonlinear inexact inner solvers is also developed.

math.OC↗

arXiv2023

Compositional Prompt Tuning with Motion Cues for Open-vocabulary Video Relation Detection

Prompt tuning with large-scale pretrained vision-language models empowers open-vocabulary predictions trained on limited base categories, e.g., object classification and detection. In this paper, we propose compositional prompt tuning with motion cues: an extended prompt tuning paradigm for compositional predictions of video data. In particular, we present Relation Prompt (RePro) for Open-vocabulary Video Visual Relation Detection (Open-VidVRD), where conventional prompt tuning is easily biased to certain subject-object combinations and motion patterns. To this end, RePro addresses the two technical challenges of Open-VidVRD: 1) the prompt tokens should respect the two different semantic roles of subject and object, and 2) the tuning should account for the diverse spatio-temporal motion patterns of the subject-object compositions. Without bells and whistles, our RePro achieves a new state-of-the-art performance on two VidVRD benchmarks of not only the base training object and predicate categories, but also the unseen ones. Extensive ablations also demonstrate the effectiveness of the proposed compositional and multi-mode design of prompts. Code is available at https://github.com/Dawn-LX/OpenVoc-VidVRD.

cs.CV↗

arXiv2023

VL-NMS: Breaking Proposal Bottlenecks in Two-Stage Visual-Language Matching

The prevailing framework for matching multimodal inputs is based on a two-stage process: 1) detecting proposals with an object detector and 2) matching text queries with proposals. Existing two-stage solutions mostly focus on the matching step. In this paper, we argue that these methods overlook an obvious \emph{mismatch} between the roles of proposals in the two stages: they generate proposals solely based on the detection confidence (i.e., query-agnostic), hoping that the proposals contain all instances mentioned in the text query (i.e., query-aware). Due to this mismatch, chances are that proposals relevant to the text query are suppressed during the filtering process, which in turn bounds the matching performance. To this end, we propose VL-NMS, which is the first method to yield query-aware proposals at the first stage. VL-NMS regards all mentioned instances as critical objects, and introduces a lightweight module to predict a score for aligning each proposal with a critical object. These scores can guide the NMS operation to filter out proposals irrelevant to the text query, increasing the recall of critical objects, resulting in a significantly improved matching performance. Since VL-NMS is agnostic to the matching step, it can be easily integrated into any state-of-the-art two-stage matching methods. We validate the effectiveness of VL-NMS on two multimodal matching tasks, namely referring expression grounding and image-text matching. Extensive ablation studies on several baselines and benchmarks consistently demonstrate the superiority of VL-NMS.

cs.CV↗

arXiv2023

Physical properties, electronic structure, and strain-tuned monolayer of the weak topological insulator RbTi3Bi5 with Kagome lattice

Kagome metals AV3Sb5 (A = K, Rb, and Cs) with a V-Kagome lattice acting as a fertile platform to investigate geometric frustration, electron correlation, superconductivity, and nontrivial band topology, have attracted tremendous attention. Here we reported the structure and properties of ATi3Bi5 (A = Rb, Cs) family with a Ti-Kagome lattice, specifically focusing on the electronic structure and nontrivial band topology of RbTi3Bi5. ATi3Bi5 (A = Rb, Cs) is found to be non-superconducting metal with strong quasi-two-dimensional feature, moderate electron correlation, and small Pauli paramagnetism. Based on first principles calculations, RbTi3Bi5 is determined to be a weak topological insulator with gapless surface states along (100) plane, and the electronic band structure along (001) plane is in great agreement with experimentally observed one. In particular, the electronic properties of the RbTi3Bi5 monolayer can be efficiently tuned by a biaxial strain according to calculation, with its lower saddle points coming from Kagome lattice approaching the Fermi level. These results highlight ATi3Bi5 (A = Rb, Cs) with Ti-Kagome lattice is a new Kagome metal to explore nontrivial band topology and exotic phases.

cond-mat.supr-con↗

arXiv2022

Network Characteristics of LEO Satellite Constellations: A Starlink-Based Measurement from End Users

Low Earth orbit Satellite Networks (LSNs) have been advocated as a key infrastructure for truly global coverage in the forthcoming 6G. This paper presents our initial measurement results and observations on the end-to-end network characteristics of Starlink, arguably the largest LSN constellation to date. Our findings confirm that LSNs are a promising solution towards ubiquitous Internet coverage over the Earth; yet, we also find that the users of Starlink experience much more dynamics in throughput and latency than terrestrial network users, and even frequent outages. Its user experiences are heavily affected by environmental factors such as terrain, solar storms, rain, clouds, and temperature, so is the power consumption. We further analyze Starlink's current bent-pipe relay strategy and its limits, particularly for cross-ocean routes. We have also explored its mobility and portability potentials, and extended our experiments from urban cities to wild remote areas that are facing distinct practical and cultural challenges.

cs.NI↗

arXiv2022

MRTNet: Multi-Resolution Temporal Network for Video Sentence Grounding

Given an untrimmed video and natural language query, video sentence grounding aims to localize the target temporal moment in the video. Existing methods mainly tackle this task by matching and aligning semantics of the descriptive sentence and video segments on a single temporal resolution, while neglecting the temporal consistency of video content in different resolutions. In this work, we propose a novel multi-resolution temporal video sentence grounding network: MRTNet, which consists of a multi-modal feature encoder, a Multi-Resolution Temporal (MRT) module, and a predictor module. MRT module is an encoder-decoder network, and output features in the decoder part are in conjunction with Transformers to predict the final start and end timestamps. Particularly, our MRT module is hot-pluggable, which means it can be seamlessly incorporated into any anchor-free models. Besides, we utilize a hybrid loss to supervise cross-modal features in MRT module for more accurate grounding in three scales: frame-level, clip-level and sequence-level. Extensive experiments on three prevalent datasets have shown the effectiveness of MRTNet.

cs.CV↗

arXiv2022

Short term prediction of demand for ride hailing services: A deep learning approach

As ride-hailing services become increasingly popular, being able to accurately predict demand for such services can help operators efficiently allocate drivers to customers, and reduce idle time, improve congestion, and enhance the passenger experience. This paper proposes UberNet, a deep learning Convolutional Neural Network for short-term prediction of demand for ride-hailing services. UberNet empploys a multivariate framework that utilises a number of temporal and spatial features that have been found in the literature to explain demand for ride-hailing services. The proposed model includes two sub-networks that aim to encode the source series of various features and decode the predicting series, respectively. To assess the performance and effectiveness of UberNet, we use 9 months of Uber pickup data in 2014 and 28 spatial and temporal features from New York City. By comparing the performance of UberNet with several other approaches, we show that the prediction quality of the model is highly competitive. Further, Ubernet's prediction performance is better when using economic, social and built environment features. This suggests that Ubernet is more naturally suited to including complex motivators in making real-time passenger demand predictions for ride-hailing services.

cs.LG↗

arXiv2022

Flux Variations of Cosmic Ray Air Showers Detected by LHAASO-KM2A During a Thunderstorm on 10 June 2021

The Large High Altitude Air Shower Observatory (LHAASO) has three sub-arrays, KM2A, WCDA and WFCTA. The flux variations of cosmic ray air showers were studied by analyzing the KM2A data during the thunderstorm on 10 June 2021. The number of shower events that meet the trigger conditions increases significantly in atmospheric electric fields, with maximum fractional increase of 20%. The variations of trigger rates (increases or decreases) are found to be strongly dependent on the primary zenith angle. The flux of secondary particles increases significantly, following a similar trend with that of the shower events. To better understand the observed behavior, Monte Carlo simulations are performed with CORSIKA and G4KM2A (a code based on GEANT4). We find that the experimental data (in saturated negative fields) are in good agreement with simulations, assuming the presence of a uniform upward electric field of 700 V/cm with a thickness of 1500 m in the atmosphere above the observation level. Due to the acceleration/deceleration and deflection by the atmospheric electric field, the number of secondary particles with energy above the detector threshold is modified, resulting in the changes in shower detection rate.

astro-ph.HE↗

arXiv2022

Sequential Transformer for End-to-End Person Search

Person Search aims to simultaneously localize and recognize a target person from realistic and uncropped gallery images. One major challenge of person search comes from the contradictory goals of the two sub-tasks, i.e., person detection focuses on finding the commonness of all persons so as to distinguish persons from the background, while person re-identification (re-ID) focuses on the differences among different persons. In this paper, we propose a novel Sequential Transformer (SeqTR) for end-to-end person search to deal with this challenge. Our SeqTR contains a detection transformer and a novel re-ID transformer that sequentially addresses detection and re-ID tasks. The re-ID transformer comprises the self-attention layer that utilizes contextual information and the cross-attention layer that learns local fine-grained discriminative features of the human body. Moreover, the re-ID transformer is shared and supervised by multi-scale features to improve the robustness of learned person representations. Extensive experiments on two widely-used person search benchmarks, CUHK-SYSU and PRW, show that our proposed SeqTR not only outperforms all existing person search methods with a 59.3% mAP on PRW but also achieves comparable performance to the state-of-the-art results with an mAP of 94.8% on CUHK-SYSU.

cs.CV↗

arXiv2022

Boundary Proposal Network for Two-Stage Natural Language Video Localization

We aim to address the problem of Natural Language Video Localization (NLVL)-localizing the video segment corresponding to a natural language description in a long and untrimmed video. State-of-the-art NLVL methods are almost in one-stage fashion, which can be typically grouped into two categories: 1) anchor-based approach: it first pre-defines a series of video segment candidates (e.g., by sliding window), and then does classification for each candidate; 2) anchor-free approach: it directly predicts the probabilities for each video frame as a boundary or intermediate frame inside the positive segment. However, both kinds of one-stage approaches have inherent drawbacks: the anchor-based approach is susceptible to the heuristic rules, further limiting the capability of handling videos with variant length. While the anchor-free approach fails to exploit the segment-level interaction thus achieving inferior results. In this paper, we propose a novel Boundary Proposal Network (BPNet), a universal two-stage framework that gets rid of the issues mentioned above. Specifically, in the first stage, BPNet utilizes an anchor-free model to generate a group of high-quality candidate video segments with their boundaries. In the second stage, a visual-language fusion layer is proposed to jointly model the multi-modal interaction between the candidate and the language query, followed by a matching score rating layer that outputs the alignment score for each candidate. We evaluate our BPNet on three challenging NLVL benchmarks (i.e., Charades-STA, TACoS and ActivityNet-Captions). Extensive experiments and ablative studies on these datasets demonstrate that the BPNet outperforms the state-of-the-art methods.

cs.CV↗

arXiv2022

Rethinking Multi-Modal Alignment in Video Question Answering from Feature and Sample Perspectives

Reasoning about causal and temporal event relations in videos is a new destination of Video Question Answering (VideoQA).The major stumbling block to achieve this purpose is the semantic gap between language and video since they are at different levels of abstraction. Existing efforts mainly focus on designing sophisticated architectures while utilizing frame- or object-level visual representations. In this paper, we reconsider the multi-modal alignment problem in VideoQA from feature and sample perspectives to achieve better performance. From the view of feature,we break down the video into trajectories and first leverage trajectory feature in VideoQA to enhance the alignment between two modalities. Moreover, we adopt a heterogeneous graph architecture and design a hierarchical framework to align both trajectory-level and frame-level visual feature with language feature. In addition, we found that VideoQA models are largely dependent on language priors and always neglect visual-language interactions. Thus, two effective yet portable training augmentation strategies are designed to strengthen the cross-modal correspondence ability of our model from the view of sample. Extensive results show that our method outperforms all the state-of-the-art models on the challenging NExT-QA benchmark, which demonstrates the effectiveness of the proposed method.

cs.CV↗

arXiv2022

No-reference Point Cloud Geometry Quality Assessment Based on Pairwise Rank Learning

Objective geometry quality assessment of point clouds is essential to evaluate the performance of a wide range of point cloud-based solutions, such as denoising, simplification, reconstruction, and watermarking. Existing point cloud quality assessment (PCQA) methods dedicate to assigning absolute quality scores to distorted point clouds. Their performance is strongly reliant on the quality and quantity of subjective ground-truth scores for training, which are challenging to gather and have been shown to be imprecise, biased, and inconsistent. Furthermore, the majority of existing objective geometry quality assessment approaches are carried out by full-reference traditional metrics. So far, point-based no-reference geometry-only quality assessment techniques have not yet been investigated. This paper presents PRL-GQA, the first pairwise learning framework for no-reference geometry-only quality assessment of point clouds, to the best of our knowledge. The proposed PRL-GQA framework employs a siamese deep architecture, which takes as input a pair of point clouds and outputs their rank order. Each siamese architecture branch is a geometry quality assessment network (GQANet), which is designed to extract multi-scale quality-aware geometric features and output a quality index for the input point cloud. Then, based on the predicted quality indexes, a pairwise rank learning module is introduced to rank the relative quality of a pair of degraded point clouds.Extensive experiments demonstrate the effectiveness of the proposed PRL-GQA framework. Furthermore, the results also show that the fine-tuned no-reference GQANet performs competitively when compared to existing full-reference geometry quality assessment metrics.

cs.MM↗

arXiv2022

Respecting Transfer Gap in Knowledge Distillation

Knowledge distillation (KD) is essentially a process of transferring a teacher model's behavior, e.g., network response, to a student model. The network response serves as additional supervision to formulate the machine domain, which uses the data collected from the human domain as a transfer set. Traditional KD methods hold an underlying assumption that the data collected in both human domain and machine domain are both independent and identically distributed (IID). We point out that this naive assumption is unrealistic and there is indeed a transfer gap between the two domains. Although the gap offers the student model external knowledge from the machine domain, the imbalanced teacher knowledge would make us incorrectly estimate how much to transfer from teacher to student per sample on the non-IID transfer set. To tackle this challenge, we propose Inverse Probability Weighting Distillation (IPWD) that estimates the propensity score of a training sample belonging to the machine domain, and assigns its inverse amount to compensate for under-represented samples. Experiments on CIFAR-100 and ImageNet demonstrate the effectiveness of IPWD for both two-stage distillation and one-stage self-distillation.

cs.CV↗