Searcharxiv⌕ Search

arXiv subjects

Tong Wu

Publications and source records attributed to Tong Wu.

At least 253 records · Page 14Linked to original sources
arXiv2023

Existence of nontrivial solutions for critical biharmonic equations with logarithmic term

In this paper, we consider the existence of nontrivial solutions to the following critical biharmonic problem with a logarithmic term \begin{equation*} \begin{cases} Δ^2 u=μΔu+λu+|u|^{2^{**}-2}u+τu\log u^2, \ \ x\inΩ, u|_{\partial Ω}=\frac{\partial u}{\partial n}|_{\partialΩ}=0, \end{cases} \end{equation*} where $μ,λ,τ\in \mathbb{R}$, $|μ|+|τ|\ne 0$, $Δ^2=ΔΔ$ denotes the iterated N-dimensional Laplacian, $Ω\subset \mathbb{R}^{N}$ is a bounded domain with smooth boundary $\partial Ω$, $2^{**}=\frac{2N}{N-4}(N\ge5)$ is the critical Sobolev exponent for the embedding $H_{0}^{2}(Ω)\hookrightarrow L^{2^{**}}(Ω)$ and $H_0^2 (Ω)$ is the closure of $C_0^ \infty (Ω)$ under the norm $|| u ||:=(\int_Ω|Δu|^2)^\frac{1}{2}$. The uncertainty of the sign of $s\log s^2$ in $(0,+\infty)$ has some interest in itself. To know which of the three terms $μΔu$, $λu$ and $τu \log u^2$ has a greater influence on the existence of nontrivial weak solutions, we prove the existence of nontrivial weak solutions to the above problem for $N\ge5$ under some assumptions of $λ, μ$ and $τ$.

math.AP↗
arXiv2023

Generation of extreme ultraviolet laser by single-photon process

To generate laser with short wavelength is a bottle-neck problem in laser technology. The extreme ultraviolet (EUV) lasers are usually produced with low efficiency by nonlinear multi-photon process. Here we show the generation of 58.4 nm laser by a single-photon-excitation related stimulated anti-Stokes Raman scattering (ASRS) with an efficiency higher than that of high harmonic generation (HHG). Utilizing microwave excited helium in the metastable 1s2s state irradiated by a resonant femto-second 2058 nm laser, we obtain pulsed 58.4 nm laser with a divergence of 1.9 mrad, suggesting its coherent character. The pulse decay time of 426 ps shorter than the lifetime of spontaneous emission also shows its stimulated property. Our results show an applicable path towards up-conversion by single-photon process to generate intense laser with wavelength shorter than 58.4 nm.

physics.optics↗
arXiv2023

Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise

In this paper, we introduce a novel dIffusion language modEl pre-training framework for text generation, which we call GENIE. GENIE is a large-scale pretrained diffusion language model that consists of an encoder and a diffusion-based decoder, which can generate text by gradually transforming a random noise sequence into a coherent text sequence. To pre-train GENIE on a large-scale language corpus, we design a new continuous paragraph denoise objective, which encourages the diffusion-decoder to reconstruct a clean text paragraph from a corrupted version, while preserving the semantic and syntactic coherence. We evaluate GENIE on four downstream text generation benchmarks, namely XSum, CNN/DailyMail, Gigaword, and CommonGen. Our experimental results show that GENIE achieves comparable performance with the state-of-the-art autoregressive models on these benchmarks, and generates more diverse text samples. The code and models of GENIE are available at https://github.com/microsoft/ProphetNet/tree/master/GENIE.

cs.CL↗
arXiv2023

Uncovering Adversarial Risks of Test-Time Adaptation

Recently, test-time adaptation (TTA) has been proposed as a promising solution for addressing distribution shifts. It allows a base model to adapt to an unforeseen distribution during inference by leveraging the information from the batch of (unlabeled) test data. However, we uncover a novel security vulnerability of TTA based on the insight that predictions on benign samples can be impacted by malicious samples in the same batch. To exploit this vulnerability, we propose Distribution Invading Attack (DIA), which injects a small fraction of malicious data into the test batch. DIA causes models using TTA to misclassify benign and unperturbed test data, providing an entirely new capability for adversaries that is infeasible in canonical machine learning pipelines. Through comprehensive evaluations, we demonstrate the high effectiveness of our attack on multiple benchmarks across six TTA methods. In response, we investigate two countermeasures to robustify the existing insecure TTA implementations, following the principle of "security by design". Together, we hope our findings can make the community aware of the utility-security tradeoffs in deploying TTA and provide valuable insights for developing robust TTA approaches.

cs.LG↗
arXiv2023

LF-checker: Machine Learning Acceleration of Bounded Model Checking for Concurrency Verification (Competition Contribution)

We describe and evaluate LF-checker, a metaverifier tool based on machine learning. It extracts multiple features of the program under test and predicts the optimal configuration (flags) of a bounded model checker with a decision tree. Our current work is specialised in concurrency verification and employs ESBMC as a back-end verification engine. In the paper, we demonstrate that LF-checker achieves better results than the default configuration of the underlying verification engine.

cs.LG↗
arXiv2022

Riemannian geometry of noncommutative super surfaces

In this paper, a Riemannian geometry of noncommutative super surfaces is developed which generalizes [4] to the super case. The notions of metric and connections on such noncommutative super surfaces are introduced and it is shown that the connections are metric-compatible and have zero torsion when the super metric is symmetric, giving rise to the corresponding super Riemann curvature. The latter also satisfies the noncommutative super analogue of the first and second Bianchi identities. We also give some examples and study them in details.

math.DG↗
arXiv2022

SLAN: Self-Locator Aided Network for Cross-Modal Understanding

Learning fine-grained interplay between vision and language allows to a more accurate understanding for VisionLanguage tasks. However, it remains challenging to extract key image regions according to the texts for semantic alignments. Most existing works are either limited by textagnostic and redundant regions obtained with the frozen detectors, or failing to scale further due to its heavy reliance on scarce grounding (gold) data to pre-train detectors. To solve these problems, we propose Self-Locator Aided Network (SLAN) for cross-modal understanding tasks without any extra gold data. SLAN consists of a region filter and a region adaptor to localize regions of interest conditioned on different texts. By aggregating cross-modal information, the region filter selects key regions and the region adaptor updates their coordinates with text guidance. With detailed region-word alignments, SLAN can be easily generalized to many downstream tasks. It achieves fairly competitive results on five cross-modal understanding tasks (e.g., 85.7% and 69.2% on COCO image-to-text and text-to-image retrieval, surpassing previous SOTA methods). SLAN also demonstrates strong zero-shot and fine-tuned transferability to two localization tasks.

cs.CV↗
arXiv2022

Human-Robot Commensality: Bite Timing Prediction for Robot-Assisted Feeding in Groups

We develop data-driven models to predict when a robot should feed during social dining scenarios. Being able to eat independently with friends and family is considered one of the most memorable and important activities for people with mobility limitations. While existing robotic systems for feeding people with mobility limitations focus on solitary dining, commensality, the act of eating together, is often the practice of choice. Sharing meals with others introduces the problem of socially appropriate bite timing for a robot, i.e. the appropriate timing for the robot to feed without disrupting the social dynamics of a shared meal. Our key insight is that bite timing strategies that take into account the delicate balance of social cues can lead to seamless interactions during robot-assisted feeding in a social dining scenario. We approach this problem by collecting a Human-Human Commensality Dataset (HHCD) containing 30 groups of three people eating together. We use this dataset to analyze human-human commensality behaviors and develop bite timing prediction models in social dining scenarios. We also transfer these models to human-robot commensality scenarios. Our user studies show that prediction improves when our algorithm uses multimodal social signaling cues between diners to model bite timing. The HHCD dataset, videos of user studies, and code are available at https://emprise.cs.cornell.edu/hrcom/

cs.RO↗
arXiv2022

Complex-Value Spatio-temporal Graph Convolutional Neural Networks and its Applications to Electric Power Systems AI

The effective representation, precessing, analysis, and visualization of large-scale structured data over graphs are gaining a lot of attention. So far most of the literature has focused on real-valued signals. However, signals are often sparse in the Fourier domain, and more informative and compact representations for them can be obtained using the complex envelope of their spectral components, as opposed to the original real-valued signals. Motivated by this fact, in this work we generalize graph convolutional neural networks (GCN) to the complex domain, deriving the theory that allows to incorporate a complex-valued graph shift operators (GSO) in the definition of graph filters (GF) and process complex-valued graph signals (GS). The theory developed can handle spatio-temporal complex network processes. We prove that complex-valued GCNs are stable with respect to perturbations of the underlying graph support, the bound of the transfer error and the bound of error propagation through multiply layers. Then we apply complex GCN to power grid state forecasting, power grid cyber-attack detection and localization.

cs.LG↗
arXiv2022

A Multiscale Simulation Approach for Germanium-Hole-Based Quantum Processor

A multiscale simulation method is developed to model a quantum dot (QD) array of germanium (Ge) holes for quantum computing. Guided by three-dimensional numerical quantum device simulations of QD structures, an analytical model of the tunnel coupling between the neighboring hole QDs is obtained. Two-qubit entangling quantum gate operations and quantum circuit characteristics of the QD array processor are then modeled. Device analysis of two-qubit Ge hole quantum gates demonstrates faster gate speed, smaller process variability, and less stringent requirement of feature size, compared to its silicon counterpart. The multiscale simulation method allows assessment of the quantum processor circuit performance from a bottom-up, physics-informed perspective. Application of the simulation method to the Ge QD array processor indicates its promising potential for preparing high-fidelity ansatz states in quantum chemistry simulations.

cs.ET↗
arXiv2022

Just Rotate it: Deploying Backdoor Attacks via Rotation Transformation

Recent works have demonstrated that deep learning models are vulnerable to backdoor poisoning attacks, where these attacks instill spurious correlations to external trigger patterns or objects (e.g., stickers, sunglasses, etc.). We find that such external trigger signals are unnecessary, as highly effective backdoors can be easily inserted using rotation-based image transformation. Our method constructs the poisoned dataset by rotating a limited amount of objects and labeling them incorrectly; once trained with it, the victim's model will make undesirable predictions during run-time inference. It exhibits a significantly high attack success rate while maintaining clean performance through comprehensive empirical studies on image classification and object detection tasks. Furthermore, we evaluate standard data augmentation techniques and four different backdoor defenses against our attack and find that none of them can serve as a consistent mitigation approach. Our attack can be easily deployed in the real world since it only requires rotating the object, as we show in both image classification and object detection applications. Overall, our work highlights a new, simple, physically realizable, and highly effective vector for backdoor attacks. Our video demo is available at https://youtu.be/6JIF8wnX34M.

cs.CV↗
arXiv2022

Multiscale Modeling of Semimetal Contact to Two-Dimensional Transition Metal Dichalcogenide Semiconductor

A multiscale simulation approach is developed to simulate the contact transport properties between semimetal to a monolayer two-dimensional (2D) transition metal dichalcogenide (TMDC) semiconductor. The results elucidate the mechanisms for low contact resistance between semimetal and TMDC semiconductor contacts from a quantum transport perspective. The simulation results compare favorably with recent experiments. Furthermore, the results show that the contact resistance of a Bismuth-MoS2 contact can be further reduced by engineering the dielectric environment and doping the TMDC material to <100 $Ω$.$μ$m. The quantum transport simulation indicates the possibility to achieve an ultrashort contact transfer length of ~1 nm, which can allow aggressive scaling of the contact size.

cond-mat.mes-hall↗
arXiv2022

Few-Shot Object Detection via Association and DIscrimination

Object detection has achieved substantial progress in the last decade. However, detecting novel classes with only few samples remains challenging, since deep learning under low data regime usually leads to a degraded feature space. Existing works employ a holistic fine-tuning paradigm to tackle this problem, where the model is first pre-trained on all base classes with abundant samples, and then it is used to carve the novel class feature space. Nonetheless, this paradigm is still imperfect. Durning fine-tuning, a novel class may implicitly leverage the knowledge of multiple base classes to construct its feature space, which induces a scattered feature space, hence violating the inter-class separability. To overcome these obstacles, we propose a two-step fine-tuning framework, Few-shot object detection via Association and DIscrimination (FADI), which builds up a discriminative feature space for each novel class with two integral steps. 1) In the association step, in contrast to implicitly leveraging multiple base classes, we construct a compact novel class feature space via explicitly imitating a specific base class feature space. Specifically, we associate each novel class with a base class according to their semantic similarity. After that, the feature space of a novel class can readily imitate the well-trained feature space of the associated base class. 2) In the discrimination step, to ensure the separability between the novel classes and associated base classes, we disentangle the classification branches for base and novel classes. To further enlarge the inter-class separability between all classes, a set-specialized margin loss is imposed. Extensive experiments on Pascal VOC and MS-COCO datasets demonstrate FADI achieves new SOTA performance, significantly improving the baseline in any shot/split by +18.7. Notably, the advantage is most announced on extremely few-shot scenarios.

cs.CV↗
arXiv2022

Affine connections on singular multiply warped products and singular twisted products

In this paper, we generalize the results in [Y. Wang: Affine connections on singular warped products. Int. J. Geom. Methods Mod. Phys. 18(5), 2150076, (2021).] to singular multiply warped products and singular twisted products. We study singular multiply warped products and singular twisted products and their curvature with the semi-symmetric metric connection and the semi-symmetric non-metric connection. We also discuss Koszul forms associated with the almost product structure and their curvature of singular multiply warped products and singular twisted products. Finally, several examples are presented to demonstrate the theoretical results.

math.DG↗
arXiv2022

Adversarial Robustness of Deep Sensor Fusion Models

We experimentally study the robustness of deep camera-LiDAR fusion architectures for 2D object detection in autonomous driving. First, we find that the fusion model is usually both more accurate, and more robust against single-source attacks than single-sensor deep neural networks. Furthermore, we show that without adversarial training, early fusion is more robust than late fusion, whereas the two perform similarly after adversarial training. However, we note that single-channel adversarial training of deep fusion is often detrimental even to robustness. Moreover, we observe cross-channel externalities, where single-channel adversarial training reduces robustness to attacks on the other channel. Additionally, we observe that the choice of adversarial model in adversarial training is critical: using attacks restricted to cars' bounding boxes is more effective in adversarial training and exhibits less significant cross-channel externalities. Finally, we find that joint-channel adversarial training helps mitigate many of the issues above, but does not significantly boost adversarial robustness.

cs.CV↗
arXiv2022

Privacy-Preserving Distributed Optimal Power Flow with Partially Homomorphic Encryption

Distribution grid agents are obliged to exchange and disclose their states explicitly to neighboring regions to enable distributed optimal power flow dispatch. However, the states contain sensitive information of individual agents, such as voltage and current measurements. These measurements can be inferred by adversaries, such as other participating agents or eavesdroppers. To address the issue, we propose a privacy-preserving distributed optimal power flow (OPF) algorithm based on partially homomorphic encryption (PHE). First of all, we exploit the alternating direction method of multipliers (ADMM) to solve the OPF in a distributed fashion. In this way, the dual update of ADMM can be encrypted by PHE. We further relax the augmented term of the primal update of ADMM with the $\ell_1$-norm regularization. In addition, we transform the relaxed ADMM with the $\ell_1$-norm regularization to a semidefinite program (SDP), and prove that this transformation is exact. The SDP can be solved locally with only the sign messages from neighboring agents, which preserves the privacy of the primal update. At last, we strictly prove the privacy preservation guarantee of the proposed algorithm. Numerical case studies validate the effectiveness and exactness of the proposed approach.

eess.SY↗