arXiv subjects

Han Wu

Publications and source records attributed to Han Wu.

At least 127 records · Page 7Linked to original sources

arXiv2023

Reversible Non-Volatile Electronic Switching in a Near Room Temperature van der Waals Ferromagnet

The ability to reversibly toggle between two distinct states in a non-volatile method is important for information storage applications. Such devices have been realized for phase-change materials, which utilizes local heating methods to toggle between a crystalline and an amorphous state with distinct electrical properties. To expand such kind of switching between two topologically distinct phases requires non-volatile switching between two crystalline phases with distinct symmetries. Here we report the observation of reversible and non-volatile switching between two stable and closely-related crystal structures with remarkably distinct electronic structures in the near room temperature van der Waals ferromagnet Fe$_{5-δ}$GeTe$_2$. From a combination of characterization techniques we show that the switching is enabled by the ordering and disordering of an Fe site vacancy that results in distinct crystalline symmetries of the two phases that can be controlled by a thermal annealing and quenching method. Furthermore, from symmetry analysis as well as first principle calculations, we provide understanding of the key distinction in the observed electronic structures of the two phases: topological nodal lines compatible with the preserved global inversion symmetry in the site-disordered phase, and flat bands resulting from quantum destructive interference on a bipartite crystaline lattice formed by the presence of the site order as well as the lifting of the topological degeneracy due to the broken inversion symmetry in the site-ordered phase. Our work not only reveals a rich variety of quantum phases emergent in the metallic van der Waals ferromagnets due to the presence of site ordering, but also demonstrates the potential of these highly tunable two-dimensional magnets for memory and spintronics applications.

cond-mat.str-el↗

arXiv2023

VCSUM: A Versatile Chinese Meeting Summarization Dataset

Compared to news and chat summarization, the development of meeting summarization is hugely decelerated by the limited data. To this end, we introduce a versatile Chinese meeting summarization dataset, dubbed VCSum, consisting of 239 real-life meetings, with a total duration of over 230 hours. We claim our dataset is versatile because we provide the annotations of topic segmentation, headlines, segmentation summaries, overall meeting summaries, and salient sentences for each meeting transcript. As such, the dataset can adapt to various summarization tasks or methods, including segmentation-based summarization, multi-granularity summarization and retrieval-then-generate summarization. Our analysis confirms the effectiveness and robustness of VCSum. We also provide a set of benchmark models regarding different downstream summarization tasks on VCSum to facilitate further research. The dataset and code will be released at https://github.com/hahahawu/VCSum.

cs.CL↗

arXiv2023

Detecting Interference in A/B Testing with Increasing Allocation

In the past decade, the technology industry has adopted online randomized controlled experiments (a.k.a. A/B testing) to guide product development and make business decisions. In practice, A/B tests are often implemented with increasing treatment allocation: the new treatment is gradually released to an increasing number of units through a sequence of randomized experiments. In scenarios such as experimenting in a social network setting or in a bipartite online marketplace, interference among units may exist, which can harm the validity of simple inference procedures. In this work, we introduce a widely applicable procedure to test for interference in A/B testing with increasing allocation. Our procedure can be implemented on top of an existing A/B testing platform with a separate flow and does not require a priori a specific interference mechanism. In particular, we introduce two permutation tests that are valid under different assumptions. Firstly, we introduce a general statistical test for interference requiring no additional assumption. Secondly, we introduce a testing procedure that is valid under a time fixed effect assumption. The testing procedure is of very low computational complexity, it is powerful, and it formalizes a heuristic algorithm implemented already in industry. We demonstrate the performance of the proposed testing procedure through simulations on synthetic data. Finally, we discuss one application at LinkedIn, where a screening step is implemented to detect potential interference in all their marketplace experiments with the proposed methods in the paper.

stat.ME↗

arXiv2023

A uniform Weyl bound for L-functions of Hilbert modular forms

We establish a Weyl-type subconvexity of $L(\tfrac{1}{2},f)$ for spherical Hilbert newforms $f$ with level ideal $\mathfrak{N}^2$, in which $\mathfrak{N}$ is required to be cube-free, and at any prime ideal $\mathfrak{p}$ with $\mathfrak{p}^2 \mid \mathfrak{N}$ the local representation generated by $f$ is not supercuspidal. The proof exploits a distributional version of Motohashi's formula over number fields developed by the first author, as well as Katz's work on hypergeometric sums over finite fields in the language of $\ell$-adic cohomology.

math.NT↗

arXiv2023

Learning Locality and Isotropy in Dialogue Modeling

Existing dialogue modeling methods have achieved promising performance on various dialogue tasks with the aid of Transformer and the large-scale pre-trained language models. However, some recent studies revealed that the context representations produced by these methods suffer the problem of anisotropy. In this paper, we find that the generated representations are also not conversational, losing the conversation structure information during the context modeling stage. To this end, we identify two properties in dialogue modeling, i.e., locality and isotropy, and present a simple method for dialogue representation calibration, namely SimDRC, to build isotropic and conversational feature spaces. Experimental results show that our approach significantly outperforms the current state-of-the-art models on three dialogue tasks across the automatic and human evaluation metrics. More in-depth analyses further confirm the effectiveness of our proposed approach.

cs.CL↗

arXiv2022

FastMIM: Expediting Masked Image Modeling Pre-training for Vision

The combination of transformers and masked image modeling (MIM) pre-training framework has shown great potential in various vision tasks. However, the pre-training computational budget is too heavy and withholds the MIM from becoming a practical training paradigm. This paper presents FastMIM, a simple and generic framework for expediting masked image modeling with the following two steps: (i) pre-training vision backbones with low-resolution input images; and (ii) reconstructing Histograms of Oriented Gradients (HOG) feature instead of original RGB values of the input images. In addition, we propose FastMIM-P to progressively enlarge the input resolution during pre-training stage to further enhance the transfer results of models with high capacity. We point out that: (i) a wide range of input resolutions in pre-training phase can lead to similar performances in fine-tuning phase and downstream tasks such as detection and segmentation; (ii) the shallow layers of encoder are more important during pre-training and discarding last several layers can speed up the training stage with no harm to fine-tuning performance; (iii) the decoder should match the size of selected network; and (iv) HOG is more stable than RGB values when resolution transfers;. Equipped with FastMIM, all kinds of vision backbones can be pre-trained in an efficient way. For example, we can achieve 83.8%/84.1% top-1 accuracy on ImageNet-1K with ViT-B/Swin-B as backbones. Compared to previous relevant approaches, we can achieve comparable or better top-1 accuracy while accelerate the training procedure by $\sim$5$\times$. Code can be found in https://github.com/ggjy/FastMIM.pytorch.

cs.CV↗

arXiv2022

Know Your Customer: Balancing Innovation and Regulation for Financial Inclusion

Financial inclusion depends on providing adjusted services for citizens with disclosed vulnerabilities. At the same time, the financial industry needs to adhere to a strict regulatory framework, which is often in conflict with the desire for inclusive, adaptive, and privacy-preserving services. In this article we study how this tension impacts the deployment of privacy-sensitive technologies aimed at financial inclusion. We conduct a qualitative study with banking experts to understand their perspectives on service development for financial inclusion. We build and demonstrate a prototype solution based on open source decentralized identifiers and verifiable credentials software and report on feedback from the banking experts on this system. The technology is promising thanks to its selective disclosure of vulnerabilities to the full control of the individual. This supports GDPR requirements, but at the same time, there is a clear tension between introducing these technologies and fulfilling other regulatory requirements, particularly with respect to 'Know Your Customer.' We consider the policy implications stemming from these tensions and provide guidelines for the further design of related technologies.

cs.CY↗

arXiv2022

Federated Learning for Tabular Data: Exploring Potential Risk to Privacy

Federated Learning (FL) has emerged as a potentially powerful privacy-preserving machine learning methodology, since it avoids exchanging data between participants, but instead exchanges model parameters. FL has traditionally been applied to image, voice and similar data, but recently it has started to draw attention from domains including financial services where the data is predominantly tabular. However, the work on tabular data has not yet considered potential attacks, in particular attacks using Generative Adversarial Networks (GANs), which have been successfully applied to FL for non-tabular data. This paper is the first to explore leakage of private data in Federated Learning systems that process tabular data. We design a Generative Adversarial Networks (GANs)-based attack model which can be deployed on a malicious client to reconstruct data and its properties from other participants. As a side-effect of considering tabular data, we are able to statistically assess the efficacy of the attack (without relying on human observation such as done for FL for images). We implement our attack model in a recently developed generic FL software framework for tabular data processing. The experimental results demonstrate the effectiveness of the proposed attack model, thus suggesting that further research is required to counter GAN-based privacy attacks.

cs.CR↗

arXiv2022

Radiation build-up and dissipation in random fiber laser

Random fiber laser (RFL) is a complex physical system that arises from the distributed amplification and the intrinsic stochasticity of the fiber scattering. There has been widespread interest in analyzing the underlying lightwave kinetics at steady state. However, the transient state, such as the RFL build-up and dissipation, is also particularly important for unfolding lightwave interaction process. Here, we investigate for the first time the RFL dynamics at transient state, and track the RFL temporal and spectral evolution theoretically and experimentally. Particularly, with the contribution of randomly distributed feedback, the build-up of RFL shows continuous Verhulst logistic growth curves without cavity-related features, which is significantly different from the step-like growth curve of conventional fiber lasers. Furthermore, the radiation build-up duration is inversely related to the pump power, and the spectral evolution of RFL undergoes two phases from spectral density increase to spectral broadening. From steady-state to pump switch-off state, the RFL output power dissipates immediately, and the remaining Stokes lightwave from the Rayleigh scattering will gradually disappear after one round-trip. This work provides new insights into the transient dynamics features of the RFL.

physics.optics↗

arXiv2022

Non-invariance of weak approximation with Brauer-Manin obstruction for surfaces

In this paper, we study the property of weak approximation with Brauer-Manin obstruction for surfaces with respect to field extensions of number fields. For any nontrivial extension of number fields L/K, assuming a conjecture of M. Stoll, we construct a smooth, projective, and geometrically connected surface over K such that it satisfies weak approximation with Brauer-Manin obstruction off all archimedean places, while its base change to L fails. Then we illustrate this construction with an explicit unconditional example.

math.NT↗

arXiv2022

Stacked conductive metal organic framework nanorods for high performance vacuum electronic devices

Metal-organic frameworks (MOFs) possessing many unique features have been utilized in several fields in recent years. However, their application in field emission (FE) vacuum electronic device is hindered by their poor electrical conductivity. Herein, a novel conductive MOF of Cu-catecholate (Cu-CAT) with the nanorod length of 200 nm and conductivity of 0.01 S/cm is grown on the graphite paper (GP). Under an applied electric field, a large number of electrons can be emitted from the nanoscale emitter tips of MOF surface to the anode. The great field emission performance of Cu-CAT@GP cold cathode film including a low turn-on field of 0.59e6 V/m and ultra-high field enhancement factor of 29622, even comparable to most carbon-based materials that have been widely investigated in FE studies, is achieved in this work. Meanwhile, Cu-CAT@GP film has a good electrical stability with a current attenuation of 9.4% in two hours. The findings reveal the cathode film fabricated by conductive MOF can be a promising candidate of cold electron source for vacuum electronic applications.

physics.app-ph↗

arXiv2022

Interpretable Personalized Experimentation

Black-box heterogeneous treatment effect (HTE) models are increasingly being used to create personalized policies that assign individuals to their optimal treatments. However, they are difficult to understand, and can be burdensome to maintain in a production environment. In this paper, we present a scalable, interpretable personalized experimentation system, implemented and deployed in production at Meta. The system works in a multiple treatment, multiple outcome setting typical at Meta to: (1) learn explanations for black-box HTE models; (2) generate interpretable personalized policies. We evaluate the methods used in the system on publicly available data and Meta use cases, and discuss lessons learnt during the development of the system.

cs.LG↗

arXiv2022

Explicit subconvexity for $\mathrm{GL}_2$

We make the subconvex exponent for $\mathrm{GL}_2$ cuspidal representation in the work of Michel \& Venkatesh explicit. The result depends on an effective dependence on the `fixed' $\mathrm{GL}_2$ representation in our former work on the subconvex bounds for twists by Hecke characters, which in turn depends on the $\mathrm{L}^4$-norm of the test function.

math.NT↗

arXiv2022

Partial Likelihood Thompson Sampling

We consider the problem of deciding how best to target and prioritize existing vaccines that may offer protection against new variants of an infectious disease. Sequential experiments are a promising approach; however, challenges due to delayed feedback and the overall ebb and flow of disease prevalence make available methods inapplicable for this task. We present a method, partial likelihood Thompson sampling, that can handle these challenges. Our method involves running Thompson sampling with belief updates determined by partial likelihood each time we observe an event. To test our approach, we ran a semi-synthetic experiment based on 200 days of COVID-19 infection data in the US.

stat.ME↗

arXiv2022

CMT: Convolutional Neural Networks Meet Vision Transformers

Vision transformers have been successfully applied to image recognition tasks due to their ability to capture long-range dependencies within an image. However, there are still gaps in both performance and computational cost between transformers and existing convolutional neural networks (CNNs). In this paper, we aim to address this issue and develop a network that can outperform not only the canonical transformers, but also the high-performance convolutional models. We propose a new transformer based hybrid network by taking advantage of transformers to capture long-range dependencies, and of CNNs to model local features. Furthermore, we scale it to obtain a family of models, called CMTs, obtaining much better accuracy and efficiency than previous convolution and transformer based models. In particular, our CMT-S achieves 83.5% top-1 accuracy on ImageNet, while being 14x and 2x smaller on FLOPs than the existing DeiT and EfficientNet, respectively. The proposed CMT-S also generalizes well on CIFAR10 (99.2%), CIFAR100 (91.7%), Flowers (98.7%), and other challenging vision datasets such as COCO (44.3% mAP), with considerably less computational cost.

cs.CV↗

arXiv2022

On Motohashi's formula

We complement and offer a new perspective of the proof of a Motohashi-type formula relating the fourth moment of $L$-functions for $\mathrm{GL}_1$ with the third moment of $L$-functions for $\mathrm{GL}_2$ over number fields, studied earlier by Michel-Venkatesh and Nelson. Our main tool is a new type of pre-trace formula with test functions on $\mathrm{M}_2(\mathbb{A})$ instead of $\mathrm{GL}_2(\mathbb{A})$, on whose spectral side the matrix coefficients are replaced by the standard Godement-Jacquet zeta integrals. This is also a generalization of Bruggeman-Motohashi's other proof of Motohashi's formula. We give a variation of our method in the case of division quaternion algebras instead of $\mathrm{M}_2$, yielding a new spectral reciprocity, for which we are not sure if it is within the period formalism given by Michel-Venkatesh. We also indicate a further possible generalization, which seems to be beyond what the period method can offer.

math.NT↗

arXiv2022

CBlockSim: A Modular High-Performance Blockchain Simulator

Blockchain has attracted much attention from both academia and industry since emerging in 2008. Due to the inconvenience of the deployment of large-scale blockchains, blockchain simulators are used to facilitate blockchain design and implementation. We evaluate state-of-the-art simulators applied to both Bitcoin and Ethereum and find that they suffer from low performance and scalability which are significant limitations. To build a more general and faster blockchain simulator, we extend an existing blockchain simulator, i.e. BlockSim. We add a network module integrated with a network topology generation algorithm and a block propagation algorithm to generate a realistic blockchain network and simulate the block propagation efficiently. We design a binary transaction pool structure and migrate BlockSim from Python to C++ so that bitwise operations can be used to accelerate the simulation and reduce memory usage. Moreover, we modularize the simulator based on five primary blockchain processes. Significant blockchain elements including consensus protocols (PoW and PoS), information propagation algorithms (Gossip) and finalization rules (Longest rule and GHOST rule) are implemented in individual modules and can be combined flexibly to simulate different types of blockchains. Experiments demonstrate that the new simulator reduces the simulation time by an order of magnitude and improves scalability, enabling us to simulate more than ten thousand nodes, roughly the size of the Bitcoin and Ethereum networks. Two typical use cases are proposed to investigate network-related issues which are not covered by most other simulators.

cs.DC↗

arXiv2022

Bias of Root Numbers for Hilbert Newforms of Cubic Level

We give a general formula of the bias of root numbers for Hilbert modular newforms of cubic level. Explicit calculation is given when the base field is $\mathbb{Q}, \mathbb{Q}(\sqrt{2}), \mathbb{Q}(\sqrt{5})$ and the level is the cube of certain rational integers. This complements a previous result of the second author and extends the bias phenomenon to the number fields. Our method is based on Jacquet-Zagier's trace formula, and the explicit calculation works generally for all real quadratic fields of narrow class number one and for rational cubic levels.

math.NT↗