arXiv subjects

Hao Zhang

Publications and source records attributed to Hao Zhang.

At least 901 records · Page 50Linked to original sources

arXiv2023

Spin-charge separation and unconventional superconductivity in \textit{t}-\textit{J} model on honeycomb lattice

The physical nature of doped Mott-insulator has been intensively studied for more than three decades. It is well known that the single band Hubbard model or $t$-$J$ model on the bipartite lattice is the simplest model to describe a doped Mott insulator. Unfortunately, the key mechanism of superconductivity in these toy models is still under debate. In this paper, we propose a new mechanism for the $d+id$-wave superconductivity (SC) that occurs in the small-doping region of the honeycomb lattice $t$-$J$ model based on the Grassmann tensor product state numerical simulation and spin-charge separation formulation. Moreover, in the presence of anti-ferromagnetic order, a continuum effective field theory for holons is developed near half-filling. It reveals the competition between repulsive and attractive holon interactions induced by spinon fluctuations and gauge fluctuations, respectively. At a large value of $t/J$, the repulsive interaction dominates, leading to the non-Fermi liquid like behavior; while in a moderate range of $t/J$, the attractive interaction dominates, leading to the SC order. Possible experimental detection of spin-charge separation phenomena is also discussed.

cond-mat.str-el↗

arXiv2022

Langevin dynamics of generalized spins as SU($N$) coherent states

Classical models of spin systems traditionally retain only the dipole moments, but a quantum spin state will frequently have additional structure. Spins of magnitude $S$ have $N=2S+1$ levels. Alternatively, the spin state is fully characterized by a set of $N^{2}-1$ local physical observables, which we interpret as generalized spin components. For example, a spin with $S=1$ has three dipole components and five quadrupole components. These components evolve under a generalization of the classical Landau-Lifshitz dynamics, which can be extended with noise and damping terms. In this paper, we reformulate the dynamical equations of motion as a Langevin dynamics of SU($N$) coherent states in the Schrödinger picture. This viewpoint is especially useful as the basis for an efficient numerical method to sample spin configurations in thermal equilibrium and to simulate the relaxation and driven motion of topological solitons. To illustrate the approach, we simulate a non-equilibrium relaxation process that creates CP$^{2}$ Skyrmions, which are topological defects with both dipole and quadrupole character.

cond-mat.str-el↗

arXiv2022

Improved Long-Form Spoken Language Translation with Large Language Models

A challenge in spoken language translation is that plenty of spoken content is long-form, but short units are necessary for obtaining high-quality translations. To address this mismatch, we fine-tune a general-purpose, large language model to split long ASR transcripts into segments that can be independently translated so as to maximize the overall translation quality. We compare to several segmentation strategies and find that our approach improves BLEU score on three languages by an average of 2.7 BLEU overall compared to an automatic punctuation baseline. Further, we demonstrate the effectiveness of two constrained decoding strategies to improve well-formedness of the model output from above 99% to 100%.

cs.CL↗

arXiv2022

Towards Robust Calculation of index-$k$ Saddle Point: Iterative Proximal-Minimization and Differential Game Model

Saddle point with a given Morse index on a potential energy surface is an important object related to energy landscape in physics and chemistry. Efficient numerical methods based on iterative minimization formulation have been proposed in the forms of the sequence of minimization subproblems or the continuous dynamics. We here present a differential game interpretation of this formulation and theoretically investigate the Nash equilibrium of the proposed game and the original saddle point on potential energy surface. To define this differential game, a new proximal function growing faster than quadratic is introduced to the cost function in the game and a robust Iterative Proximal-Minimization algorithm (IPM) is then derived to compute the saddle points. We prove that the Nash equilibrium of the game is exactly the saddle point in concern and show that the new algorithm is more robust than the previous iterative minimization algorithm without proximity, while the same convergence rate and the computational cost still hold. A two dimensional problem and the Cahn-Hillard problem are tested to demonstrate this numerical advantage.

math.OC↗

arXiv2022

Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation

In this paper we present Mask DINO, a unified object detection and segmentation framework. Mask DINO extends DINO (DETR with Improved Denoising Anchor Boxes) by adding a mask prediction branch which supports all image segmentation tasks (instance, panoptic, and semantic). It makes use of the query embeddings from DINO to dot-product a high-resolution pixel embedding map to predict a set of binary masks. Some key components in DINO are extended for segmentation through a shared architecture and training process. Mask DINO is simple, efficient, and scalable, and it can benefit from joint large-scale detection and segmentation datasets. Our experiments show that Mask DINO significantly outperforms all existing specialized segmentation methods, both on a ResNet-50 backbone and a pre-trained model with SwinL backbone. Notably, Mask DINO establishes the best results to date on instance segmentation (54.5 AP on COCO), panoptic segmentation (59.4 PQ on COCO), and semantic segmentation (60.8 mIoU on ADE20K) among models under one billion parameters. Code is available at \url{https://github.com/IDEACVR/MaskDINO}.

cs.CV↗

arXiv2022

Exposing new scalars hiding behind the Higgs boson

It is possible that there is another scalar hiding behind the known 125 GeV Higgs boson. If the hidden scalar exhibits a CP property different from the Higgs boson, it can be exposed in the di-Higgs production at the high-luminosity large hadron collider and future colliders.

hep-ph↗

arXiv2022

DN-DETR: Accelerate DETR Training by Introducing Query DeNoising

We present in this paper a novel denoising training method to speedup DETR (DEtection TRansformer) training and offer a deepened understanding of the slow convergence issue of DETR-like methods. We show that the slow convergence results from the instability of bipartite graph matching which causes inconsistent optimization goals in early training stages. To address this issue, except for the Hungarian loss, our method additionally feeds ground-truth bounding boxes with noises into Transformer decoder and trains the model to reconstruct the original boxes, which effectively reduces the bipartite graph matching difficulty and leads to a faster convergence. Our method is universal and can be easily plugged into any DETR-like methods by adding dozens of lines of code to achieve a remarkable improvement. As a result, our DN-DETR results in a remarkable improvement ($+1.9$AP) under the same setting and achieves the best result (AP $43.4$ and $48.6$ with $12$ and $50$ epochs of training respectively) among DETR-like methods with ResNet-$50$ backbone. Compared with the baseline under the same setting, DN-DETR achieves comparable performance with $50\%$ training epochs. Code is available at \url{https://github.com/FengLi-ust/DN-DETR}.

cs.CV↗

arXiv2022

Proximity effect in PbTe-Pb hybrid nanowire Josephson junctions

Semiconductor-superconductor hybrid nanowires are a leading material platform for the realization of Majorana zero modes. The semiconductors in previous studies are dominantly InAs or InSb. In this work, we show the induced superconductivity in a PbTe nanowire epitaxially coupled to a superconductor Pb. The Josephson junction device based on this hybrid reveals a gate-tunable supercurrent in the open regime and a hard superconducting gap in the tunneling regime. By demonstrating the superconducting proximity effect, our result can enable Majorana searches and other applications like gate-tunable qubits in a new semiconductor system.

cond-mat.mes-hall↗

arXiv2022

SAC-GAN: Structure-Aware Image Composition

We introduce an end-to-end learning framework for image-to-image composition, aiming to plausibly compose an object represented as a cropped patch from an object image into a background scene image. As our approach emphasizes more on semantic and structural coherence of the composed images, rather than their pixel-level RGB accuracies, we tailor the input and output of our network with structure-aware features and design our network losses accordingly, with ground truth established in a self-supervised setting through the object cropping. Specifically, our network takes the semantic layout features from the input scene image, features encoded from the edges and silhouette in the input object patch, as well as a latent code as inputs, and generates a 2D spatial affine transform defining the translation and scaling of the object patch. The learned parameters are further fed into a differentiable spatial transformer network to transform the object patch into the target image, where our model is trained adversarially using an affine transform discriminator and a layout discriminator. We evaluate our network, coined SAC-GAN, for various image composition scenarios in terms of quality, composability, and generalizability of the composite images. Comparisons are made to state-of-the-art alternatives, including Instance Insertion, ST-GAN, CompGAN and PlaceNet, confirming superiority of our method.

cs.CV↗

arXiv2022

Deterministic synthesis of phase pure 2D perovskites via progressive transformation of layer thickness

Two-dimensional (2D) halide perovskites have emerged as semiconductor platforms for realizing efficient and durable optoelectronic devices. However, the reproducible synthesis of 2D perovskite crystals with desired layer thickness (or n value) greater than 2, has been an enduring challenge due to the lack of kinetic control (temperature, time, stoichiometry) for each layer thickness. Here, we demonstrate a novel method term as the kinetically controlled space confinement (KCSC) for the deterministic growth of phase pure Ruddlesden-Popper (RP) and Dion-Jacobson (DJ) 2D perovskites. The phase-pure growth was achieved by progressively increasing the temperature (fixed time) or the crystallization time (fixed temperature), which allowed for an acute control of the crystallization kinetics. We also observe a systematic transformation from a lower n-value to n=3, 4, 5, 6 in 2D perovskites. In-situ photoluminescence spectroscopy and imaging suggest that the progressive transformation from lower to higher n-value occurs via intercalation of excess precursor ions. These experiments enabled the development of a machine learning assisted multi-parameter phase diagram, which predicts the growth of 2D phase with a specific n-value.

cond-mat.mtrl-sci↗

arXiv2022

DQ-DETR: Dual Query Detection Transformer for Phrase Extraction and Grounding

In this paper, we study the problem of visual grounding by considering both phrase extraction and grounding (PEG). In contrast to the previous phrase-known-at-test setting, PEG requires a model to extract phrases from text and locate objects from images simultaneously, which is a more practical setting in real applications. As phrase extraction can be regarded as a $1$D text segmentation problem, we formulate PEG as a dual detection problem and propose a novel DQ-DETR model, which introduces dual queries to probe different features from image and text for object prediction and phrase mask prediction. Each pair of dual queries is designed to have shared positional parts but different content parts. Such a design effectively alleviates the difficulty of modality alignment between image and text (in contrast to a single query design) and empowers Transformer decoder to leverage phrase mask-guided attention to improve performance. To evaluate the performance of PEG, we also propose a new metric CMAP (cross-modal average precision), analogous to the AP metric in object detection. The new metric overcomes the ambiguity of Recall@1 in many-box-to-one-phrase cases in phrase grounding. As a result, our PEG pre-trained DQ-DETR establishes new state-of-the-art results on all visual grounding benchmarks with a ResNet-101 backbone. For example, it achieves $91.04\%$ and $83.51\%$ in terms of recall rate on RefCOCO testA and testB with a ResNet-101 backbone. Code will be availabl at \url{https://github.com/IDEA-Research/DQ-DETR}.

cs.CV↗

arXiv2022

Coordinating Cross-modal Distillation for Molecular Property Prediction

In recent years, molecular graph representation learning (GRL) has drawn much more attention in molecular property prediction (MPP) problems. The existing graph methods have demonstrated that 3D geometric information is significant for better performance in MPP. However, accurate 3D structures are often costly and time-consuming to obtain, limiting the large-scale application of GRL. It is an intuitive solution to train with 3D to 2D knowledge distillation and predict with only 2D inputs. But some challenging problems remain open for 3D to 2D distillation. One is that the 3D view is quite distinct from the 2D view, and the other is that the gradient magnitudes of atoms in distillation are discrepant and unstable due to the variable molecular size. To address these challenging problems, we exclusively propose a distillation framework that contains global molecular distillation and local atom distillation. We also provide a theoretical insight to justify how to coordinate atom and molecular information, which tackles the drawback of variable molecular size for atom information distillation. Experimental results on two popular molecular datasets demonstrate that our proposed model achieves superior performance over other methods. Specifically, on the largest MPP dataset PCQM4Mv2 served as an "ImageNet Large Scale Visual Recognition Challenge" in the field of graph ML, the proposed method achieved a 6.9% improvement compared with the best works. And we obtained fourth place with the MAE of 0.0734 on the test-challenge set for OGB-LSC 2022 Graph Regression Task. We will release the code soon.

cs.LG↗

arXiv2022

Entanglement entropy scaling of noisy random quantum circuits in two dimensions

Whether noisy quantum devices without error correction can provide quantum advantage over classical computers is a critical issue of current quantum computation. In this work, the random quantum circuits, which are used as the paradigm model to demonstrate quantum advantage, are simulated with depolarizing noise on experiment relevant two-dimensional architecture. With comprehensive numerical simulation and theoretical analysis, we find that the maximum achievable operator entanglement entropy, which indicates maximal simulation cost, has area law scaling with the system size for constant noise rate. On the other hand, we also find that the maximum achievable operator entanglement entropy has power law scaling with the noise rate for fixed system size, and the volume law scaling can be obtained only if the noise rate decreases when system size increase.

quant-ph↗

arXiv2022

Identifying Correlation in Stream of Samples

Identifying independence between two random variables or correlated given their samples has been a fundamental problem in Statistics. However, how to do so in a space-efficient way if the number of states is large is not quite well-studied. We propose a new, simple counter matrix algorithm, which utilize hash functions and a compressed counter matrix to give an unbiased estimate of the $\ell_2$ independence metric. With $\mathcal{O}(ε^{-4}\logδ^{-1})$ (very loose bound) space, we can guarantee $1\pmε$ multiplicative error with probability at least $1-δ$. We also provide a comparison of our algorithm with the state-of-the-art sketching of sketches algorithm and show that our algorithm is effective, and actually faster and at least 2 times more space-efficient.

cs.DS↗

arXiv2022

A Unified Mutual Supervision Framework for Referring Expression Segmentation and Generation

Reference Expression Segmentation (RES) and Reference Expression Generation (REG) are mutually inverse tasks that can be naturally jointly trained. Though recent work has explored such joint training, the mechanism of how RES and REG can benefit each other is still unclear. In this paper, we propose a unified mutual supervision framework that enables two tasks to improve each other. Our mutual supervision contains two directions. On the one hand, Disambiguation Supervision leverages the expression unambiguity measurement provided by RES to enhance the language generation of REG. On the other hand, Generation Supervision uses expressions automatically generated by REG to scale up the training of RES. Such unified mutual supervision effectively improves two tasks by solving their bottleneck problems. Extensive experiments show that our approach significantly outperforms all existing methods on REG and RES tasks under the same setting, and detailed ablation studies demonstrate the effectiveness of all components in our framework.

cs.CV↗

arXiv2022

FF2: A Feature Fusion Two-Stream Framework for Punctuation Restoration

To accomplish punctuation restoration, most existing methods focus on introducing extra information (e.g., part-of-speech) or addressing the class imbalance problem. Recently, large-scale transformer-based pre-trained language models (PLMS) have been utilized widely and obtained remarkable success. However, the PLMS are trained on the large dataset with marks, which may not fit well with the small dataset without marks, causing the convergence to be not ideal. In this study, we propose a Feature Fusion two-stream framework (FF2) to bridge the gap. Specifically, one stream leverages a pre-trained language model to capture the semantic feature, while another auxiliary module captures the feature at hand. We also modify the computation of multi-head attention to encourage communication among heads. Then, two features with different perspectives are aggregated to fuse information and enhance context awareness. Without additional data, the experimental results on the popular benchmark IWSLT demonstrate that FF2 achieves new SOTA performance, which verifies that our approach is effective.

cs.CL↗

arXiv2022

Discovering and Explaining the Representation Bottleneck of DNNs

This paper explores the bottleneck of feature representations of deep neural networks (DNNs), from the perspective of the complexity of interactions between input variables encoded in DNNs. To this end, we focus on the multi-order interaction between input variables, where the order represents the complexity of interactions. We discover that a DNN is more likely to encode both too simple interactions and too complex interactions, but usually fails to learn interactions of intermediate complexity. Such a phenomenon is widely shared by different DNNs for different tasks. This phenomenon indicates a cognition gap between DNNs and human beings, and we call it a representation bottleneck. We theoretically prove the underlying reason for the representation bottleneck. Furthermore, we propose a loss to encourage/penalize the learning of interactions of specific complexities, and analyze the representation capacities of interactions of different complexities.

cs.LG↗

arXiv2022

A Variational Edge Partition Model for Supervised Graph Representation Learning

Graph neural networks (GNNs), which propagate the node features through the edges and learn how to transform the aggregated features under label supervision, have achieved great success in supervised feature extraction for both node-level and graph-level classification tasks. However, GNNs typically treat the graph structure as given and ignore how the edges are formed. This paper introduces a graph generative process to model how the observed edges are generated by aggregating the node interactions over a set of overlapping node communities, each of which contributes to the edges via a logical OR mechanism. Based on this generative model, we partition each edge into the summation of multiple community-specific weighted edges and use them to define community-specific GNNs. A variational inference framework is proposed to jointly learn a GNN-based inference network that partitions the edges into different communities, these community-specific GNNs, and a GNN-based predictor that combines community-specific GNNs for the end classification task. Extensive evaluations on real-world graph datasets have verified the effectiveness of the proposed method in learning discriminative representations for both node-level and graph-level classification tasks.

stat.ML↗