arXiv subjects

Meng Tang

Publications and source records attributed to Meng Tang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Manipulating magnetization by orbital current from a light metal Ti

The orbital Hall effect, which does not rely on the spin-orbit coupling, has recently emerged as a promising mechanism for electrically manipulating magnetization in thin-film ferromagnets. Despite its potential, direct experimental observation of magnetization switching driven by orbital currents has been challenging, primarily because there is no direct exchange coupling between orbital angular momentum and local spin based magnetic moments. In this study, we present a compensated design to directly probe the contribution of orbital currents in the most promising light metal titanium (Ti), where symmetric layer structures allow zeroing out of the net spin current. By varying the thickness of the Ti layer in Ti(t)/Pt/Co/Pt/Co/Pt multilayers, we demonstrate the ability to control the magnetization switching polarity. We deduce the orbital charge conversion efficiency of the Ti layer to be approximately 0.17. These findings not only confirm the presence of the orbital Hall effect in Ti but also suggest that orbital currents may be promising candidates for developing energy-efficient magnetic devices with enhanced performance and scalability.

cond-mat.mtrl-sci↗

arXiv2025

Self-Cross Diffusion Guidance for Text-to-Image Synthesis of Similar Subjects

Diffusion models achieved unprecedented fidelity and diversity for synthesizing image, video, 3D assets, etc. However, subject mixing is an unresolved issue for diffusion-based image synthesis, particularly for synthesizing multiple similar-looking subjects. We propose Self-Cross Diffusion Guidance to penalize the overlap between cross-attention maps and the aggregated self-attention map. Compared to previous methods based on self-attention or cross-attention alone, our guidance is more effective in eliminating subject mixing. What's more, our guidance addresses subject mixing for all relevant patches beyond the most discriminant one, e.g., the beak of a bird. For each subject, we aggregate self-attention maps of patches with higher cross-attention values. Thus, the aggregated self-attention map forms a region that the whole subject attends to. Our training-free method boosts the performance of both Unet-based and Transformer-based diffusion models such as the Stable Diffusion series. We also release a similar subjects dataset (SSD), a challenging benchmark, and utilize GPT-4o for automatic and reliable evaluation. Extensive qualitative and quantitative results demonstrate the effectiveness of our self-cross diffusion guidance.

cs.CV↗

arXiv2024

ScribbleGen: Generative Data Augmentation Improves Scribble-supervised Semantic Segmentation

Recent advances in generative models, such as diffusion models, have made generating high-quality synthetic images widely accessible. Prior works have shown that training on synthetic images improves many perception tasks, such as image classification, object detection, and semantic segmentation. We are the first to explore generative data augmentations for scribble-supervised semantic segmentation. We propose ScribbleGen, a generative data augmentation method that leverages a ControlNet diffusion model conditioned on semantic scribbles to produce high-quality training data. However, naive implementations of generative data augmentations may inadvertently harm the performance of the downstream segmentor rather than improve it. We leverage classifier-free diffusion guidance to enforce class consistency and introduce encode ratios to trade off data diversity for data realism. Using the guidance scale and encode ratio, we can generate a spectrum of high-quality training images. We propose multiple augmentation schemes and find that these schemes significantly impact model performance, especially in the low-data regime. Our framework further reduces the gap between the performance of scribble-supervised segmentation and that of fully-supervised segmentation. We also show that our framework significantly improves segmentation performance on small datasets, even surpassing fully-supervised segmentation. The code is available at https://github.com/mengtang-lab/scribblegen.

cs.CV↗

arXiv2024

Training Class-Imbalanced Diffusion Model Via Overlap Optimization

Diffusion models have made significant advances recently in high-quality image synthesis and related tasks. However, diffusion models trained on real-world datasets, which often follow long-tailed distributions, yield inferior fidelity for tail classes. Deep generative models, including diffusion models, are biased towards classes with abundant training images. To address the observed appearance overlap between synthesized images of rare classes and tail classes, we propose a method based on contrastive learning to minimize the overlap between distributions of synthetic images for different classes. We show variants of our probabilistic contrastive learning method can be applied to any class conditional diffusion model. We show significant improvement in image synthesis using our loss for multiple datasets with long-tailed distribution. Extensive experimental results demonstrate that the proposed method can effectively handle imbalanced data for diffusion-based generation and classification models. Our code and datasets will be publicly available at https://github.com/yanliang3612/DiffROP.

cs.CV↗

arXiv2024

Magnetic Field Gated and Current Controlled Spintronic Mem-transistor Neuron -based Spiking Neural Networks

Spintronic devices, such as the domain walls and skyrmions, have shown significant potential for applications in energy-efficient data storage and beyond CMOS computing architectures. In recent years, spiking neural networks have shown more bio-plausibility. Based on the magnetic multilayer spintronic devices, we demonstrate the magnetic field-gated Leaky integrate and fire neuron characteristics for the spiking neural network applications. The LIF characteristics are controlled by the current pulses, which drive the domain wall, and an external magnetic field is used as the bias to tune the firing properties of the neuron. Thus, the device works like a gate-controlled LIF neuron, acting like a spintronic Mem-Transistor device. We develop a LIF neuron model based on the measured characteristics to show the device integration in the system-level SNNs. We extend the study and propose a scaled version of the demonstrated device with a multilayer spintronic domain wall magnetic tunnel junction as a LIF neuron. using the combination of SOT and the variation of the demagnetization energy across the thin film, the modified leaky integrate and fire LIF neuron characteristics are realized in the proposed devices. The neuron device characteristics are modeled as the modified LIF neuron model. Finally, we integrate the measured and simulated neuron models in the 3-layer spiking neural network and convolutional spiking neural network CSNN framework to test these spiking neuron models for classification of the MNIST and FMNIST datasets. In both architectures, the network achieves classification accuracy above 96%. Considering the good system-level performance, mem-transistor properties, and promise for scalability. The presented devices show an excellent properties for neuromorphic computing applications.

physics.app-ph↗

arXiv2023

Latent Space Editing in Transformer-Based Flow Matching

This paper strives for image editing via generative models. Flow Matching is an emerging generative modeling technique that offers the advantage of simple and efficient training. Simultaneously, a new transformer-based U-ViT has recently been proposed to replace the commonly used UNet for better scalability and performance in generative modeling. Hence, Flow Matching with a transformer backbone offers the potential for scalable and high-quality generative modeling, but their latent structure and editing ability are as of yet unknown. Hence, we adopt this setting and explore how to edit images through latent space manipulation. We introduce an editing space, which we call $u$-space, that can be manipulated in a controllable, accumulative, and composable manner. Additionally, we propose a tailored sampling solution to enable sampling with the more efficient adaptive step-size ODE solvers. Lastly, we put forth a straightforward yet powerful method for achieving fine-grained and nuanced editing using text prompts. Our framework is simple and efficient, all while being highly effective at editing images while preserving the essence of the original content. Our code will be publicly available at https://taohu.me/lfm/

cs.CV↗

arXiv2022

Divide and Conquer: Text Semantic Matching with Disentangled Keywords and Intents

Text semantic matching is a fundamental task that has been widely used in various scenarios, such as community question answering, information retrieval, and recommendation. Most state-of-the-art matching models, e.g., BERT, directly perform text comparison by processing each word uniformly. However, a query sentence generally comprises content that calls for different levels of matching granularity. Specifically, keywords represent factual information such as action, entity, and event that should be strictly matched, while intents convey abstract concepts and ideas that can be paraphrased into various expressions. In this work, we propose a simple yet effective training strategy for text semantic matching in a divide-and-conquer manner by disentangling keywords from intents. Our approach can be easily combined with pre-trained language models (PLM) without influencing their inference efficiency, achieving stable performance improvements against a wide range of PLMs on three benchmarks.

cs.CL↗

arXiv2021

Deep-learning-based coupled flow-geomechanics surrogate model for CO$_2$ sequestration

A deep-learning-based surrogate model capable of predicting flow and geomechanical responses in CO2 storage operations is presented and applied. The 3D recurrent R-U-Net model combines deep convolutional and recurrent neural networks to capture the spatial distribution and temporal evolution of saturation, pressure and surface displacement fields. The method is trained using high-fidelity simulation results for 2000 storage-aquifer realizations characterized by multi-Gaussian porosity and log-permeability fields. These numerical solutions are expensive because the domain that must be considered for the coupled problem includes not only the storage aquifer but also a surrounding region, overburden and bedrock. The surrogate model is trained to predict the 3D CO2 saturation and pressure fields in the storage aquifer, and 2D displacement maps at the Earth's surface. Detailed comparisons between surrogate model and full-order simulation results for new (test-case) storage-aquifer realizations are presented. The saturation, pressure and surface displacement fields provided by the surrogate model display a high degree of accuracy, both for individual test-case realizations and for ensemble statistics. Finally, the the recurrent R-U-Net surrogate model is applied with a rejection sampling procedure for data assimilation. Although the observations consist of only a small number of surface displacement measurements, significant uncertainty reduction in pressure buildup at the top of the storage aquifer (caprock) is achieved.

cs.LG↗

arXiv2021

Efficient field-free perpendicular magnetization switching by a magnetic spin Hall effect

Current induced spin-orbit torques driven by the conventional spin Hall effect are widely used to manipulate the magnetization. This approach, however, is nondeterministic and inefficient for the switching of magnets with perpendicular magnetic anisotropy that are demanded by the high-density magnetic storage and memory devices. Here, we demonstrate that this limitation can be overcome by exploiting a magnetic spin Hall effect in noncollinear antiferromagnets, such as Mn3Sn. The magnetic group symmetry of Mn3Sn allows generation of the out-of-plane spin current carrying spin polarization induced by an in-plane charge current. This spin current drives an out-of-plane anti-damping torque providing deterministic switching of perpendicular magnetization of an adjacent Ni/Co multilayer. Compared to the conventional spin-orbit torque devices, the observed switching does not need any external magnetic field and requires much lower current density. Our results demonstrate great prospects of exploiting the magnetic spin Hall effect in noncollinear antiferromagnets for low-power spintronics.

cond-mat.mtrl-sci↗

arXiv2020

Scannerless non-line-of-sight three dimensional imaging with a 32x32 SPAD array

We develop a scannerless non-line-of-sight three dimensional imaging system based on a commercial 32x32 SPAD camera combined with a 70 ps pulsed laser. In our experiment, 1024 time histograms can be achieved synchronously in 3s with an average time resolution of about 165 ps. The result with filtered back projection shows a discernable reconstruction while the result using virtual wave field demonstrates a better quality similar to the ones created by earlier scanning imaging systems with single pixel SPAD. Comparatively, our system has large potential advantages in frame frequency, power requirements, compactness and robustness. The research results will pave a path for scannerless non-line-of-sight three dimensional imaging application.

eess.IV↗

arXiv2020

Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow

Data assimilation in subsurface flow systems is challenging due to the large number of flow simulations often required, and by the need to preserve geological realism in the calibrated (posterior) models. In this work we present a deep-learning-based surrogate model for two-phase flow in 3D subsurface formations. This surrogate model, a 3D recurrent residual U-Net (referred to as recurrent R-U-Net), consists of 3D convolutional and recurrent (convLSTM) neural networks, designed to capture the spatial-temporal information associated with dynamic subsurface flow systems. A CNN-PCA procedure (convolutional neural network post-processing of principal component analysis) for parameterizing complex 3D geomodels is also described. This approach represents a simplified version of a recently developed supervised-learning-based CNN-PCA framework. The recurrent R-U-Net is trained on the simulated dynamic 3D saturation and pressure fields for a set of random `channelized' geomodels (generated using 3D CNN-PCA). Detailed flow predictions demonstrate that the recurrent R-U-Net surrogate model provides accurate results for dynamic states and well responses for new geological realizations, along with accurate flow statistics for an ensemble of new geomodels. The 3D recurrent R-U-Net and CNN-PCA procedures are then used in combination for a challenging data assimilation problem involving a channelized system. Two different algorithms, namely rejection sampling and an ensemble-based method, are successfully applied. The overall methodology described in this paper may enable the assessment and refinement of data assimilation procedures for a range of realistic and challenging subsurface flow problems.

physics.comp-ph↗

arXiv2020

FroDO: From Detections to 3D Objects

Object-oriented maps are important for scene understanding since they jointly capture geometry and semantics, allow individual instantiation and meaningful reasoning about objects. We introduce FroDO, a method for accurate 3D reconstruction of object instances from RGB video that infers object location, pose and shape in a coarse-to-fine manner. Key to FroDO is to embed object shapes in a novel learnt space that allows seamless switching between sparse point cloud and dense DeepSDF decoding. Given an input sequence of localized RGB frames, FroDO first aggregates 2D detections to instantiate a category-aware 3D bounding box per object. A shape code is regressed using an encoder network before optimizing shape and pose further under the learnt shape priors using sparse and dense shape representations. The optimization uses multi-view geometric, photometric and silhouette losses. We evaluate on real-world datasets, including Pix3D, Redwood-OS, and ScanNet, for single-view, multi-view, and multi-object reconstruction.

cs.CV↗

arXiv2019

Angular characterization of spin-orbit torque and thermoelectric effects

Arising from the interplay between charge, spin and orbital of electrons, spin-orbit torque (SOT) has attracted immense interest in the past decade. Despite vast progress, the existing quantification methods of SOT still have their respective restrictions on the magnetic anisotropy, the entanglement between SOT effective fields, and the artifacts from the thermal gradient and the planar Hall effect, etc. Thus, accurately characterizing SOT across diverse samples remains as a critical need. In this work, with the aim of removing the afore-mentioned restrictions, thus enabling the universal SOT quantification, we report the characterization of the sign and amplitude of SOT by angular measurements. We first validate the applicability of our angular characterization in a perpendicularly magnetized Pt/Co-Ni heterostructure by showing excellent agreements to the results of conventional quantification methods. Remarkably, the thermoelectric effects, i.e., the anomalous Nernst effect (ANE) arising from the temperature gradient can be self-consistently disentangled and quantified from the field dependence of the angular characterization. The superiority of this angular characterization has been further demonstrated in a Cu/CoTb/Cu sample with large ANE but negligible SOT, and in a Pt/Co-Ni sample with weak perpendicular magnetic anisotropy (PMA), for which the conventional quantification methods are not applicable and even yield fatal error. By providing a comprehensive and versatile way to characterize SOT and thermoelectric effects in diverse heterostructures, our results pave the important foundation for the spin-orbitronic study as well as the interdisciplinary research of thermal spintronic.

cond-mat.mtrl-sci↗

arXiv2019

Multiphase flow prediction with deep neural networks

This paper proposes a deep neural network approach for predicting multiphase flow in heterogeneous domains with high computational efficiency. The deep neural network model is able to handle permeability heterogeneity in high dimensional systems, and can learn the interplay of viscous, gravity, and capillary forces from small data sets. Using the example of carbon dioxide (CO2) storage, we demonstrate that the model can generate highly accurate predictions of a CO2 saturation distribution given a permeability field, injection duration, injection rate, and injection location. The trained neural network model has an excellent ability to interpolate and to a limited extent, the ability to extrapolate beyond the training data ranges. To improve the prediction accuracy when the neural network model needs to extrapolate, we propose a transfer learning (fine-tuning) procedure that can quickly teach the neural network model new information without going through massive data collection and retraining. Based on this trained neural network model, a web-based tool is provided that allows users to perform CO2-water multiphase flow calculations online. With the tools provided in this paper, the deep neural network approach can provide a computationally efficient substitute for repetitive forward multiphase flow simulations, which can be adopted to the context of history matching and uncertainty quantification.

cs.LG↗

arXiv2019

A deep-learning-based surrogate model for data assimilation in dynamic subsurface flow problems

A deep-learning-based surrogate model is developed and applied for predicting dynamic subsurface flow in channelized geological models. The surrogate model is based on deep convolutional and recurrent neural network architectures, specifically a residual U-Net and a convolutional long short term memory recurrent network. Training samples entail global pressure and saturation maps, at a series of time steps, generated by simulating oil-water flow in many (1500 in our case) realizations of a 2D channelized system. After training, the `recurrent R-U-Net' surrogate model is shown to be capable of accurately predicting dynamic pressure and saturation maps and well rates (e.g., time-varying oil and water rates at production wells) for new geological realizations. Assessments demonstrating high surrogate-model accuracy are presented for an individual geological realization and for an ensemble of 500 test geomodels. The surrogate model is then used for the challenging problem of data assimilation (history matching) in a channelized system. For this study, posterior reservoir models are generated using the randomized maximum likelihood method, with the permeability field represented using the recently developed CNN-PCA parameterization. The flow responses required during the data assimilation procedure are provided by the recurrent R-U-Net. The overall approach is shown to lead to substantial reduction in prediction uncertainty. High-fidelity numerical simulation results for the posterior geomodels (generated by the surrogate-based data assimilation procedure) are shown to be in essential agreement with the recurrent R-U-Net predictions. The accuracy and dramatic speedup provided by the surrogate model suggest that it may eventually enable the application of more formal posterior sampling methods in realistic problems.

cs.LG↗

arXiv2019

Learning Compressed Sentence Representations for On-Device Text Processing

Vector representations of sentences, trained on massive text corpora, are widely used as generic sentence embeddings across a variety of NLP problems. The learned representations are generally assumed to be continuous and real-valued, giving rise to a large memory footprint and slow retrieval speed, which hinders their applicability to low-resource (memory and computation) platforms, such as mobile devices. In this paper, we propose four different strategies to transform continuous and generic sentence embeddings into a binarized form, while preserving their rich semantic information. The introduced methods are evaluated across a wide range of downstream tasks, where the binarized sentence embeddings are demonstrated to degrade performance by only about 2% relative to their continuous counterparts, while reducing the storage requirement by over 98%. Moreover, with the learned binary representations, the semantic relatedness of two sentences can be evaluated by simply calculating their Hamming distance, which is more computational efficient compared with the inner product operation between continuous embeddings. Detailed analysis and case study further validate the effectiveness of proposed methods.

cs.CL↗

arXiv2019

Beyond Gradient Descent for Regularized Segmentation Losses

The simplicity of gradient descent (GD) made it the default method for training ever-deeper and complex neural networks. Both loss functions and architectures are often explicitly tuned to be amenable to this basic local optimization. In the context of weakly-supervised CNN segmentation, we demonstrate a well-motivated loss function where an alternative optimizer (ADM) achieves the state-of-the-art while GD performs poorly. Interestingly, GD obtains its best result for a "smoother" tuning of the loss function. The results are consistent across different network architectures. Our loss is motivated by well-understood MRF/CRF regularization models in "shallow" segmentation and their known global solvers. Our work suggests that network design/training should pay more attention to optimization methods.

cs.LG↗

arXiv2019

Constrained-CNN losses for weakly supervised segmentation

Weakly-supervised learning based on, e.g., partially labelled images or image-tags, is currently attracting significant attention in CNN segmentation as it can mitigate the need for full and laborious pixel/voxel annotations. Enforcing high-order (global) inequality constraints on the network output (for instance, to constrain the size of the target region) can leverage unlabeled data, guiding the training process with domain-specific knowledge. Inequality constraints are very flexible because they do not assume exact prior knowledge. However, constrained Lagrangian dual optimization has been largely avoided in deep networks, mainly for computational tractability reasons. To the best of our knowledge, the method of [Pathak et al., 2015] is the only prior work that addresses deep CNNs with linear constraints in weakly supervised segmentation. It uses the constraints to synthesize fully-labeled training masks (proposals) from weak labels, mimicking full supervision and facilitating dual optimization. We propose to introduce a differentiable penalty, which enforces inequality constraints directly in the loss function, avoiding expensive Lagrangian dual iterates and proposal generation. From constrained-optimization perspective, our simple penalty-based approach is not optimal as there is no guarantee that the constraints are satisfied. However, surprisingly, it yields substantially better results than the Lagrangian-based constrained CNNs in [Pathak et al., 2015], while reducing the computational demand for training. By annotating only a small fraction of the pixels, the proposed approach can reach a level of segmentation performance that is comparable to full supervision on three separate tasks. While our experiments focused on basic linear constraints such as the target-region size and image tags, our framework can be easily extended to other non-linear constraints.

cs.CV↗