arXiv subjects

Jing Xu

Publications and source records attributed to Jing Xu.

At least 181 records · Page 10Linked to original sources

arXiv2023

Training Models to Generate, Recognize, and Reframe Unhelpful Thoughts

Many cognitive approaches to well-being, such as recognizing and reframing unhelpful thoughts, have received considerable empirical support over the past decades, yet still lack truly widespread adoption in self-help format. A barrier to that adoption is a lack of adequately specific and diverse dedicated practice material. This work examines whether current language models can be leveraged to both produce a virtually unlimited quantity of practice material illustrating standard unhelpful thought patterns matching specific given contexts, and generate suitable positive reframing proposals. We propose PATTERNREFRAME, a novel dataset of about 10k examples of thoughts containing unhelpful thought patterns conditioned on a given persona, accompanied by about 27k positive reframes. By using this dataset to train and/or evaluate current models, we show that existing models can already be powerful tools to help generate an abundance of tailored practice material and hypotheses, with no or minimal additional model training required.

cs.CL↗

arXiv2023

HODINet: High-Order Discrepant Interaction Network for RGB-D Salient Object Detection

RGB-D salient object detection (SOD) aims to detect the prominent regions by jointly modeling RGB and depth information. Most RGB-D SOD methods apply the same type of backbones and fusion modules to identically learn the multimodality and multistage features. However, these features contribute differently to the final saliency results, which raises two issues: 1) how to model discrepant characteristics of RGB images and depth maps; 2) how to fuse these cross-modality features in different stages. In this paper, we propose a high-order discrepant interaction network (HODINet) for RGB-D SOD. Concretely, we first employ transformer-based and CNN-based architectures as backbones to encode RGB and depth features, respectively. Then, the high-order representations are delicately extracted and embedded into spatial and channel attentions for cross-modality feature fusion in different stages. Specifically, we design a high-order spatial fusion (HOSF) module and a high-order channel fusion (HOCF) module to fuse features of the first two and the last two stages, respectively. Besides, a cascaded pyramid reconstruction network is adopted to progressively decode the fused features in a top-down pathway. Extensive experiments are conducted on seven widely used datasets to demonstrate the effectiveness of the proposed approach. We achieve competitive performance against 24 state-of-the-art methods under four evaluation metrics.

cs.CV↗

arXiv2023

Searching for the Fakes: Efficient Neural Architecture Search for General Face Forgery Detection

As the saying goes, "seeing is believing". However, with the development of digital face editing tools, we can no longer trust what we can see. Although face forgery detection has made promising progress, most current methods are designed manually by human experts, which is labor-consuming. In this paper, we develop an end-to-end framework based on neural architecture search (NAS) for deepfake detection, which can automatically design network architectures without human intervention. First, a forgery-oriented search space is created to choose appropriate operations for this task. Second, we propose a novel performance estimation metric, which guides the search process to select more general models. The cross-dataset search is also considered to develop more general architectures. Eventually, we connect the cells in a cascaded pyramid way for final forgery classification. Compared with state-of-the-art networks artificially designed, our method achieves competitive performance in both in-dataset and cross-dataset scenarios.

cs.CV↗

arXiv2023

Improving Open Language Models by Learning from Organic Interactions

We present BlenderBot 3x, an update on the conversational model BlenderBot 3, which is now trained using organic conversation and feedback data from participating users of the system in order to improve both its skills and safety. We are publicly releasing the participating de-identified interaction data for use by the research community, in order to spur further progress. Training models with organic data is challenging because interactions with people "in the wild" include both high quality conversations and feedback, as well as adversarial and toxic behavior. We study techniques that enable learning from helpful teachers while avoiding learning from people who are trying to trick the model into unhelpful or toxic responses. BlenderBot 3x is both preferred in conversation to BlenderBot 3, and is shown to produce safer responses in challenging situations. While our current models are still far from perfect, we believe further improvement can be achieved by continued use of the techniques explored in this work.

cs.CL↗

arXiv2023

Quantifying the Variability Collapse of Neural Networks

Recent studies empirically demonstrate the positive relationship between the transferability of neural networks and the within-class variation of the last layer features. The recently discovered Neural Collapse (NC) phenomenon provides a new perspective of understanding such last layer geometry of neural networks. In this paper, we propose a novel metric, named Variability Collapse Index (VCI), to quantify the variability collapse phenomenon in the NC paradigm. The VCI metric is well-motivated and intrinsically related to the linear probing loss on the last layer features. Moreover, it enjoys desired theoretical and empirical properties, including invariance under invertible linear transformations and numerical stability, that distinguishes it from previous metrics. Our experiments verify that VCI is indicative of the variability collapse and the transferability of pretrained neural networks.

cs.LG↗

arXiv2023

A Closer Look at Few-shot Classification Again

Few-shot classification consists of a training phase where a model is learned on a relatively large dataset and an adaptation phase where the learned model is adapted to previously-unseen tasks with limited labeled samples. In this paper, we empirically prove that the training algorithm and the adaptation algorithm can be completely disentangled, which allows algorithm analysis and design to be done individually for each phase. Our meta-analysis for each phase reveals several interesting insights that may help better understand key aspects of few-shot classification and connections with other fields such as visual representation learning and transfer learning. We hope the insights and research challenges revealed in this paper can inspire future work in related directions. Code and pre-trained models (in PyTorch) are available at https://github.com/Frankluox/CloserLookAgainFewShot.

cs.LG↗

arXiv2023

Angle-based SLAM on 5G mmWave Systems: Design, Implementation, and Measurement

Simultaneous localization and mapping (SLAM) is a key technology that provides user equipment (UE) tracking and environment mapping services, enabling the deep integration of sensing and communication. The millimeter-wave (mmWave) communication, with its larger bandwidths and antenna arrays, inherently facilitates more accurate delay and angle measurements than sub-6 GHz communication, thereby providing opportunities for SLAM. However, none of the existing works have realized the SLAM function under the 5G New Radio (NR) standard due to specification and hardware constraints. In this study, we investigate how 5G mmWave communication systems can achieve situational awareness without changing the transceiver architecture and 5G NR standard. We implement 28 GHz mmWave transceivers that deploy OFDM-based 5G NR waveform with 160 MHz channel bandwidth, and we realize beam management following the 5G NR. Furthermore, we develop an efficient successive cancellation-based angle extraction approach to obtain angles of arrival and departure from the reference signal received power measurements. On the basis of angle measurements, we propose an angle-only SLAM algorithm to track UE and map features in the radio environment. Thorough experiments and ray tracing-based computer simulations verify that the proposed angle-based SLAM can achieve sub-meter level localization and mapping accuracy with a single base station and without the requirement of strict time synchronization. Our experiments also reveal many propagation properties critical to the success of SLAM in 5G mmWave communication systems.

cs.IT↗

arXiv2023

Site-testing at the Muztagh-ata Site.V. Nighttime Cloud Amount during the Last Five Years

The clarity of nights is the major factor that should be carefully considered for optical/infrared astronomical observatories in site-testing campaigns. Cloud coverage is directly related to the amount of time available for scientific observations at observatories. In this article, we report on the results of detailed night-time cloud statistics and continuous observing me derived from ground-based all-sky cameras at the Muztagh-ata site from 2017 to 2021. Results obtained from acquisition data show that the proportion of the annual observing me at the Muztagh-ata site is 65%, and the best period with the least cloud coverage and longer continuous observing time is from September to February. We made a comparison of the monthly mean observing nights obtained from our all-sky cameras and CLARA dataset, results show that the discrepancy between them may depend on the cloud top heights. On average, this site can provide 175 clear nights and 169 nights with at least 4 hours of continuous observing time per year.

astro-ph.IM↗

arXiv2023

SoK: A Systematic Evaluation of Backdoor Trigger Characteristics in Image Classification

Deep learning achieves outstanding results in many machine learning tasks. Nevertheless, it is vulnerable to backdoor attacks that modify the training set to embed a secret functionality in the trained model. The modified training samples have a secret property, i. e., a trigger. At inference time, the secret functionality is activated when the input contains the trigger, while the model functions correctly in other cases. While there are many known backdoor attacks (and defenses), deploying a stealthy attack is still far from trivial. Successfully creating backdoor triggers depends on numerous parameters. Unfortunately, research has not yet determined which parameters contribute most to the attack performance. This paper systematically analyzes the most relevant parameters for the backdoor attacks, i.e., trigger size, position, color, and poisoning rate. Using transfer learning, which is very common in computer vision, we evaluate the attack on state-of-the-art models (ResNet, VGG, AlexNet, and GoogLeNet) and datasets (MNIST, CIFAR10, and TinyImageNet). Our attacks cover the majority of backdoor settings in research, providing concrete directions for future works. Our code is publicly available to facilitate the reproducibility of our results.

cs.CV↗

arXiv2023

Rethinking the Trigger-injecting Position in Graph Backdoor Attack

Backdoor attacks have been demonstrated as a security threat for machine learning models. Traditional backdoor attacks intend to inject backdoor functionality into the model such that the backdoored model will perform abnormally on inputs with predefined backdoor triggers and still retain state-of-the-art performance on the clean inputs. While there are already some works on backdoor attacks on Graph Neural Networks (GNNs), the backdoor trigger in the graph domain is mostly injected into random positions of the sample. There is no work analyzing and explaining the backdoor attack performance when injecting triggers into the most important or least important area in the sample, which we refer to as trigger-injecting strategies MIAS and LIAS, respectively. Our results show that, generally, LIAS performs better, and the differences between the LIAS and MIAS performance can be significant. Furthermore, we explain these two strategies' similar (better) attack performance through explanation techniques, which results in a further understanding of backdoor attacks in GNNs.

cs.LG↗

arXiv2023

On the similarity of restriction of the operator to an invariant subspace

Let $M_{z}$ be the multiplication operator on the Bergman space and $M_{I}$ denote the restriction of $M_{z}$ to an invariant subspace $I$. A question raised by K. Zhu is that when are two restriction operators $M_{I}$ and $M_{J}$ are similar? In this note, we give some sufficient conditions of this problem in a general case.

math.FA↗

arXiv2023

Dialogue State Distillation Network with Inter-slot Contrastive Learning for Dialogue State Tracking

In task-oriented dialogue systems, Dialogue State Tracking (DST) aims to extract users' intentions from the dialogue history. Currently, most existing approaches suffer from error propagation and are unable to dynamically select relevant information when utilizing previous dialogue states. Moreover, the relations between the updates of different slots provide vital clues for DST. However, the existing approaches rely only on predefined graphs to indirectly capture the relations. In this paper, we propose a Dialogue State Distillation Network (DSDN) to utilize relevant information of previous dialogue states and migrate the gap of utilization between training and testing. Thus, it can dynamically exploit previous dialogue states and avoid introducing error propagation simultaneously. Further, we propose an inter-slot contrastive learning loss to effectively capture the slot co-update relations from dialogue context. Experiments are conducted on the widely used MultiWOZ 2.0 and MultiWOZ 2.1 datasets. The experimental results show that our proposed model achieves the state-of-the-art performance for DST.

cs.CL↗

arXiv2023

Sim2Real$^2$: Actively Building Explicit Physics Model for Precise Articulated Object Manipulation

Accurately manipulating articulated objects is a challenging yet important task for real robot applications. In this paper, we present a novel framework called Sim2Real$^2$ to enable the robot to manipulate an unseen articulated object to the desired state precisely in the real world with no human demonstrations. We leverage recent advances in physics simulation and learning-based perception to build the interactive explicit physics model of the object and use it to plan a long-horizon manipulation trajectory to accomplish the task. However, the interactive model cannot be correctly estimated from a static observation. Therefore, we learn to predict the object affordance from a single-frame point cloud, control the robot to actively interact with the object with a one-step action, and capture another point cloud. Further, the physics model is constructed from the two point clouds. Experimental results show that our framework achieves about 70% manipulations with <30% relative error for common articulated objects, and 30% manipulations for difficult objects. Our proposed framework also enables advanced manipulation strategies, such as manipulating with different tools. Code and videos are available on our project webpage: https://ttimelord.github.io/Sim2Real2-site/

cs.RO↗

arXiv2023

Dynamic Graph Neural Network with Adaptive Edge Attributes for Air Quality Predictions

Air quality prediction is a typical spatio-temporal modeling problem, which always uses different components to handle spatial and temporal dependencies in complex systems separately. Previous models based on time series analysis and Recurrent Neural Network (RNN) methods have only modeled time series while ignoring spatial information. Previous GCNs-based methods usually require providing spatial correlation graph structure of observation sites in advance. The correlations among these sites and their strengths are usually calculated using prior information. However, due to the limitations of human cognition, limited prior information cannot reflect the real station-related structure or bring more effective information for accurate prediction. To this end, we propose a novel Dynamic Graph Neural Network with Adaptive Edge Attributes (DGN-AEA) on the message passing network, which generates the adaptive bidirected dynamic graph by learning the edge attributes as model parameters. Unlike prior information to establish edges, our method can obtain adaptive edge information through end-to-end training without any prior information. Thus reduced the complexity of the problem. Besides, the hidden structural information between the stations can be obtained as model by-products, which can help make some subsequent decision-making analyses. Experimental results show that our model received state-of-the-art performance than other baselines.

cs.LG↗

arXiv2023

Geometry of holomorphic vector bundles and similarity of commuting operator tuples

In this paper, a new criterion for the similarity of commuting tuples of operators on Hilbert spaces is introduced. As an application, we obtain a geometric similarity invariant of tuples in the Cowen-Douglas class which gives a partial answer to a question raised by R.G. Douglas about the similarity of quasi-free Hilbert modules. Moreover, a new subclass of commuting tuples of Cowen-Douglas class is obtained.

math.FA↗

arXiv2023

Close the Optical Sensing Domain Gap by Physics-Grounded Active Stereo Sensor Simulation

In this paper, we focus on the simulation of active stereovision depth sensors, which are popular in both academic and industry communities. Inspired by the underlying mechanism of the sensors, we designed a fully physics-grounded simulation pipeline that includes material acquisition, ray-tracing-based infrared (IR) image rendering, IR noise simulation, and depth estimation. The pipeline is able to generate depth maps with material-dependent error patterns similar to a real depth sensor in real time. We conduct real experiments to show that perception algorithms and reinforcement learning policies trained in our simulation platform could transfer well to the real-world test cases without any fine-tuning. Furthermore, due to the high degree of realism of this simulation, our depth sensor simulator can be used as a convenient testbed to evaluate the algorithm performance in the real world, which will largely reduce the human effort in developing robotic algorithms. The entire pipeline has been integrated into the SAPIEN simulator and is open-sourced to promote the research of vision and robotics communities.

cs.RO↗

arXiv2022

Unreliability of two-band model analysis of magnetoresistivities in unveiling temperature-driven Lifshitz transition

Recently, anomalies in the temperature dependences of the carrier density and/or mobility derived from analysis of the magnetoresistivities using the conventional two-band model have been used to unveil intriguing temperature-induced Lifshitz transitions in various materials. For instance, two temperature-driven Lifshitz transitions were inferred to exist in the Dirac nodal-line semimetal ZrSiSe, based on two-band model analysis of the Hall magnetoconductivities where the second band exhibits a change in the carrier type from holes to electrons when the temperature decreases below T = 106 K and a dip is observed in the mobility versus temperature curve at T = 80 K. Here, we revisit the experiments and two-band model analysis on ZrSiSe. We show that the anomalies in the second band may be spurious, because the first band dominates the Hall magnetoconductivities at T > 80 K, making the carrier type and mobility obtained for the second band from the two-band model analysis unreliable. That is, care must be taken in interpreting these anomalies as evidences for temperature-driven Lifshitz transitions. Our skepticism on the existence of such phase transitions in ZrSiSe is further supported by the validation of the Kohler's rule for magnetoresistances at temperatures below 180 K. This work showcases potential issues in interpreting anomalies in the temperature dependence of the carrier density and mobility derived from the analysis of magnetoconductivities or magnetoresistivities using the conventional two-band model.

cond-mat.mtrl-sci↗

arXiv2022

TriPINet: Tripartite Progressive Integration Network for Image Manipulation Localization

Image manipulation localization aims at distinguishing forged regions from the whole test image. Although many outstanding prior arts have been proposed for this task, there are still two issues that need to be further studied: 1) how to fuse diverse types of features with forgery clues; 2) how to progressively integrate multistage features for better localization performance. In this paper, we propose a tripartite progressive integration network (TriPINet) for end-to-end image manipulation localization. First, we extract both visual perception information, e.g., RGB input images, and visual imperceptible features, e.g., frequency and noise traces for forensic feature learning. Second, we develop a guided cross-modality dual-attention (gCMDA) module to fuse different types of forged clues. Third, we design a set of progressive integration squeeze-and-excitation (PI-SE) modules to improve localization performance by appropriately incorporating multiscale features in the decoder. Extensive experiments are conducted to compare our method with state-of-the-art image forensics approaches. The proposed TriPINet obtains competitive results on several benchmark datasets.

cs.CV↗