arXiv subjects

Shu Zhao

Publications and source records attributed to Shu Zhao.

At least 37 records · Page 2Linked to original sources

arXiv2024

Reconstruct before Query: Continual Missing Modality Learning with Decomposed Prompt Collaboration

Pre-trained large multi-modal models (LMMs) exploit fine-tuning to adapt diverse user applications. Nevertheless, fine-tuning may face challenges due to deactivated sensors (e.g., cameras turned off for privacy or technical issues), yielding modality-incomplete data and leading to inconsistency in training data and the data for inference. Additionally, continuous training leads to catastrophic forgetting, diluting the knowledge in pre-trained LMMs. To overcome these challenges, we introduce a novel task, Continual Missing Modality Learning (CMML), to investigate how models can generalize when data of certain modalities is missing during continual fine-tuning. Our preliminary benchmarks reveal that existing methods suffer from a significant performance drop in CMML, even with the aid of advanced continual learning techniques. Therefore, we devise a framework termed Reconstruct before Query (RebQ). It decomposes prompts into modality-specific ones and breaks them into components stored in pools accessible via a key-query mechanism, which facilitates ParameterEfficient Fine-Tuning and enhances knowledge transferability for subsequent tasks. Meanwhile, our RebQ leverages extensive multi-modal knowledge from pre-trained LMMs to reconstruct the data of missing modality. Comprehensive experiments demonstrate that RebQ effectively reconstructs the missing modality information and retains pre-trained knowledge. Specifically, compared with the baseline, RebQ improves average precision from 20.00 to 50.92 and decreases average forgetting from 75.95 to 8.56. Code and datasets are available on https://github.com/Tree-Shu-Zhao/RebQ.pytorch

cs.CV↗

arXiv2024

Hyperbolic Hierarchical Knowledge Graph Embeddings for Link Prediction in Low Dimensions

Knowledge graph embeddings (KGE) have been validated as powerful methods for inferring missing links in knowledge graphs (KGs) that they typically map entities into Euclidean space and treat relations as transformations of entities. Recently, some Euclidean KGE methods have been enhanced to model semantic hierarchies commonly found in KGs, improving the performance of link prediction. To embed hierarchical data, hyperbolic space has emerged as a promising alternative to traditional Euclidean space, offering high fidelity and lower memory consumption. Unlike Euclidean, hyperbolic space provides countless curvatures to choose from. However, it is difficult for existing hyperbolic KGE methods to obtain the optimal curvature settings manually, thereby limiting their ability to effectively model semantic hierarchies. To address this limitation, we propose a novel KGE model called $\textbf{Hyp}$erbolic $\textbf{H}$ierarchical $\textbf{KGE}$ (HypHKGE). This model introduces attention-based learnable curvatures for hyperbolic space, which helps preserve rich semantic hierarchies. Furthermore, to utilize the preserved hierarchies for inferring missing links, we define hyperbolic hierarchical transformations based on the theory of hyperbolic geometry, including both inter-level and intra-level modeling. Experiments demonstrate the effectiveness of the proposed HypHKGE model on the three benchmark datasets (WN18RR, FB15K-237, and YAGO3-10). The source code will be publicly released at https://github.com/wjzheng96/HypHKGE.

cs.LG↗

arXiv2024

Why are hyperbolic neural networks effective? A study on hierarchical representation capability

Hyperbolic Neural Networks (HNNs), operating in hyperbolic space, have been widely applied in recent years, motivated by the existence of an optimal embedding in hyperbolic space that can preserve data hierarchical relationships (termed Hierarchical Representation Capability, HRC) more accurately than Euclidean space. However, there is no evidence to suggest that HNNs can achieve this theoretical optimal embedding, leading to much research being built on flawed motivations. In this paper, we propose a benchmark for evaluating HRC and conduct a comprehensive analysis of why HNNs are effective through large-scale experiments. Inspired by the analysis results, we propose several pre-training strategies to enhance HRC and improve the performance of downstream tasks, further validating the reliability of the analysis. Experiments show that HNNs cannot achieve the theoretical optimal embedding. The HRC is significantly affected by the optimization objectives and hierarchical structures, and enhancing HRC through pre-training strategies can significantly improve the performance of HNNs.

cs.LG↗

arXiv2023

Chirality-induced spin splitting in 1D InSeI

Spin-orbit coupling in chiral materials can induce chirality-dependent spin splitting, enabling electrical manipulation of spin polarization. Here, we use first-principles calculations to investigate the electronic states of chiral one-dimensional (1D) semiconductor InSeI, which has two enantiomorphic configurations with left- and right-handedness. We find that opposite spin states exist in the left- and right-handed 1D InSeI with significant spin splitting and spin-momentum collinear locking. Although the spin states at the conduction band minimum (CBM) and valence band maximum (VBM) of 1D InSeI are both nearly degenerate, a direct-to-indirect bandgap transition occurs when a moderate tensile strain ($\sim$4%) is applied along the 1D chain direction, leading to a sizable spin splitting ($\sim$0.11 eV) at the CBM. These findings indicate that 1D InSeI is a promising material for chiral spintronics.

cond-mat.mtrl-sci↗

arXiv2023

Less is More: Toward Zero-Shot Local Scene Graph Generation via Foundation Models

Humans inherently recognize objects via selective visual perception, transform specific regions from the visual field into structured symbolic knowledge, and reason their relationships among regions based on the allocation of limited attention resources in line with humans' goals. While it is intuitive for humans, contemporary perception systems falter in extracting structural information due to the intricate cognitive abilities and commonsense knowledge required. To fill this gap, we present a new task called Local Scene Graph Generation. Distinct from the conventional scene graph generation task, which encompasses generating all objects and relationships in an image, our proposed task aims to abstract pertinent structural information with partial objects and their relationships for boosting downstream tasks that demand advanced comprehension and reasoning capabilities. Correspondingly, we introduce zEro-shot Local scEne GrAph geNeraTion (ELEGANT), a framework harnessing foundation models renowned for their powerful perception and commonsense reasoning, where collaboration and information communication among foundation models yield superior outcomes and realize zero-shot local scene graph generation without requiring labeled supervision. Furthermore, we propose a novel open-ended evaluation metric, Entity-level CLIPScorE (ECLIPSE), surpassing previous closed-set evaluation metrics by transcending their limited label space, offering a broader assessment. Experiment results show that our approach markedly outperforms baselines in the open-ended evaluation setting, and it also achieves a significant performance boost of up to 24.58% over prior methods in the close-set setting, demonstrating the effectiveness and powerful reasoning ability of our proposed framework.

cs.CV↗

arXiv2023

NEUCORE: Neural Concept Reasoning for Composed Image Retrieval

Composed image retrieval which combines a reference image and a text modifier to identify the desired target image is a challenging task, and requires the model to comprehend both vision and language modalities and their interactions. Existing approaches focus on holistic multi-modal interaction modeling, and ignore the composed and complimentary property between the reference image and text modifier. In order to better utilize the complementarity of multi-modal inputs for effective information fusion and retrieval, we move the multi-modal understanding to fine-granularity at concept-level, and learn the multi-modal concept alignment to identify the visual location in reference or target images corresponding to text modifier. Toward the end, we propose a NEUral COncept REasoning (NEUCORE) model which incorporates multi-modal concept alignment and progressive multimodal fusion over aligned concepts. Specifically, considering that text modifier may refer to semantic concepts not existing in the reference image and requiring to be added into the target image, we learn the multi-modal concept alignment between the text modifier and the concatenation of reference and target images, under multiple-instance learning framework with image and sentence level weak supervision. Furthermore, based on aligned concepts, to form discriminative fusion features of the input modalities for accurate target image retrieval, we propose a progressive fusion strategy with unified execution architecture instantiated by the attended language semantic concepts. Our proposed approach is evaluated on three datasets and achieves state-of-the-art results.

cs.CV↗

arXiv2023

Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Driving

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations; 2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features; 3) Camera-LiDAR fusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models.

cs.CV↗

arXiv2023

Unraveling the unusually high electrical conductivity of the delafossite metal PdCoO$_2$

The prototypical delafossite metal PdCoO$_2$ has been the subject of intense interest for hosting exotic transport properties. Using first-principles transport calculations and theoretical modeling, we reveal that the high electrical conductivity of PdCoO$_2$ at room temperature originates from the contributions of both high Fermi velocities, enabled by Pd $4d_{z^2}-5s$ hybridization, and exceptionally weak electron-phonon coupling, which leads to a coupling strength ($λ=0.057$) that is nearly an order of magnitude smaller than those of common metals. The abnormally weak electron-phonon coupling in PdCoO$_2$ results from a low electronic density of states at the Fermi level, as well as the large and strongly facetted Fermi surface with suppressed Umklapp electron-phonon matrix elements. We anticipate that our work will inform the design of unconventional metals with superior transport properties.

cond-mat.mtrl-sci↗

arXiv2023

Are Intermediate Layers and Labels Really Necessary? A General Language Model Distillation Method

The large scale of pre-trained language models poses a challenge for their deployment on various devices, with a growing emphasis on methods to compress these models, particularly knowledge distillation. However, current knowledge distillation methods rely on the model's intermediate layer features and the golden labels (also called hard labels), which usually require aligned model architecture and enough labeled data respectively. Moreover, the parameters of vocabulary are usually neglected in existing methods. To address these problems, we propose a general language model distillation (GLMD) method that performs two-stage word prediction distillation and vocabulary compression, which is simple and surprisingly shows extremely strong performance. Specifically, GLMD supports more general application scenarios by eliminating the constraints of dimension and structure between models and the need for labeled datasets through the absence of intermediate layers and golden labels. Meanwhile, based on the long-tailed distribution of word frequencies in the data, GLMD designs a strategy of vocabulary compression through decreasing vocabulary size instead of dimensionality. Experimental results show that our method outperforms 25 state-of-the-art methods on the SuperGLUE benchmark, achieving an average score that surpasses the best method by 3%.

cs.CL↗

arXiv2023

GKD: A General Knowledge Distillation Framework for Large-scale Pre-trained Language Model

Currently, the reduction in the parameter scale of large-scale pre-trained language models (PLMs) through knowledge distillation has greatly facilitated their widespread deployment on various devices. However, the deployment of knowledge distillation systems faces great challenges in real-world industrial-strength applications, which require the use of complex distillation methods on even larger-scale PLMs (over 10B), limited by memory on GPUs and the switching of methods. To overcome these challenges, we propose GKD, a general knowledge distillation framework that supports distillation on larger-scale PLMs using various distillation methods. With GKD, developers can build larger distillation models on memory-limited GPUs and easily switch and combine different distillation methods within a single framework. Experimental results show that GKD can support the distillation of at least 100B-scale PLMs and 25 mainstream methods on 8 NVIDIA A100 (40GB) GPUs.

cs.CL↗

arXiv2022

Automatic Analysis of Human Body Representations in Western Art

The way the human body is depicted in classical and modern paintings is relevant for art historical analyses. Each artist has certain themes and concerns, resulting in different poses being used more heavily than others. In this paper, we propose a computer vision pipeline to analyse human pose and representations in paintings, which can be used for specific artists or periods. Specifically, we combine two pose estimation approaches (OpenPose and DensePose, respectively) and introduce methods to deal with occlusion and perspective issues. For normalisation, we map the detected poses and contours to Leonardo da Vinci's Vitruvian Man, the classical depiction of body proportions. We propose a visualisation approach for illustrating the articulation of joints in a set of paintings. Combined with a hierarchical clustering of poses, our approach reveals common and uncommon poses used by artists. Our approach improves over purely skeleton based analyses of human body in paintings.

cs.CV↗

arXiv2022

Curvature-Controlled Band Alignment Transitions in 1D van der Waals Heterostructures

One-dimensional (1D) van der Waals (vdW) heterostructures, formed between coaxial nanotubes of transition metal dichalcogenides (TMDCs), have emerged as a new area of endeavor in nanoscience. A key to designing and engineering the properties of such 1D vdW heterostructures lies on understanding the band alignment of coaxial nanotubes in the heterostructures. However, how curvature, tube diameters, and intertube coupling affect the band-edge levels and band alignment of TMDC nanotubes in 1D vdW heterostructures remains unknown. Here, through comprehensive first-principles calculations and analyses, we establish a complete framework of band alignment in 1D vdW heterostructures of TMDC nanotubes. We reveal that, as the diameter of a TMDC nanotube decreases, the combined effects of curvature-induced flexoelectricity and intrinsic circumferential tensile strain cause a rapid and continuous lowering of the conduction band minimum (CBM), whereas the valence band maximum (VBM) exhibits an initial lowering before rising, which originates from a change in the orbital character of the VBM. The transition in the orbital character of VBM also leads to direct-to-indirect bandgap transition in small-diameter armchair and chiral nanotubes, as well as photoluminescence quenching in zigzag nanotubes. As individual TMDC nanotubes form coaxial 1D vdW heterostructures, the effect of intertube coupling via flexovoltage effect can result in a transition of intertube band alignment from Type II to Type I in multiple heterostructural systems, including large-diameter MoSe$_2$@WS$_2$, MoTe$_2$@MoSe$_2$, and MoTe$_2$@WS$_2$ heterostructures. These results lay down a foundation for the rational design of 1D vdW heterostructures.

cond-mat.mtrl-sci↗

arXiv2022

Aesthetic Visual Question Answering of Photographs

Aesthetic assessment of images can be categorized into two main forms: numerical assessment and language assessment. Aesthetics caption of photographs is the only task of aesthetic language assessment that has been addressed. In this paper, we propose a new task of aesthetic language assessment: aesthetic visual question and answering (AVQA) of images. If we give a question of images aesthetics, model can predict the answer. We use images from \textit{www.flickr.com}. The objective QA pairs are generated by the proposed aesthetic attributes analysis algorithms. Moreover, we introduce subjective QA pairs that are converted from aesthetic numerical labels and sentiment analysis from large-scale pre-train models. We build the first aesthetic visual question answering dataset, AesVQA, that contains 72,168 high-quality images and 324,756 pairs of aesthetic questions. Two methods for adjusting the data distribution have been proposed and proved to improve the accuracy of existing models. This is the first work that both addresses the task of aesthetic VQA and introduces subjectiveness into VQA tasks. The experimental results reveal that our methods outperform other VQA models on this new task.

cs.CV↗

arXiv2022

Attribute Controllable Beautiful Caucasian Face Generation by Aesthetics Driven Reinforcement Learning

In recent years, image generation has made great strides in improving the quality of images, producing high-fidelity ones. Also, quite recently, there are architecture designs, which enable GAN to unsupervisedly learn the semantic attributes represented in different layers. However, there is still a lack of research on generating face images more consistent with human aesthetics. Based on EigenGAN [He et al., ICCV 2021], we build the techniques of reinforcement learning into the generator of EigenGAN. The agent tries to figure out how to alter the semantic attributes of the generated human faces towards more preferable ones. To accomplish this, we trained an aesthetics scoring model that can conduct facial beauty prediction. We also can utilize this scoring model to analyze the correlation between face attributes and aesthetics scores. Empirically, using off-the-shelf techniques from reinforcement learning would not work well. So instead, we present a new variant incorporating the ingredients emerging in the reinforcement learning communities in recent years. Compared to the original generated images, the adjusted ones show clear distinctions concerning various attributes. Experimental results using the MindSpore, show the effectiveness of the proposed method. Altered facial images are commonly more attractive, with significantly improved aesthetic levels.

cs.CV↗

arXiv2022

Giant modulation of the electron mobility in semiconductor Bi$_2$O$_2$Se via incipient ferroelectric phase transition

High-mobility layered semiconductors have the potential to enable the next-generation electronics and computing. This paper demonstrates that the ultrahigh electron mobility observed in the layered semiconductor Bi$_2$O$_2$Se originates from an incipient ferroelectric transition that endows the material with a robust protection against mobility degradation by Coulomb scattering. Based on first-principles calculations of electron-phonon interaction and ionized impurity scattering, it is shown that the electron mobility of Bi$_2$O$_2$Se can reach 10$^4$ to 10$^6$ cm$^2$V$^{-1}$s$^{-1}$ over a wide range of realistic doping concentrations. Furthermore, a small elastic strain of 1.7% can drive the material toward a unique interlayer ferroelectric transition, resulting in a large increase in the dielectric permittivity and a giant enhancement of the low-temperature electron mobility by more than an order of magnitude. These results establish a new route to realize high-mobility layered semiconductors via phase and dielectric engineering.

cond-mat.mtrl-sci↗

arXiv2022

Coherence-Based Distributed Document Representation Learning for Scientific Documents

Distributed document representation is one of the basic problems in natural language processing. Currently distributed document representation methods mainly consider the context information of words or sentences. These methods do not take into account the coherence of the document as a whole, e.g., a relation between the paper title and abstract, headline and description, or adjacent bodies in the document. The coherence shows whether a document is meaningful, both logically and syntactically, especially in scientific documents (papers or patents, etc.). In this paper, we propose a coupled text pair embedding (CTPE) model to learn the representation of scientific documents, which maintains the coherence of the document with coupled text pairs formed by segmenting the document. First, we divide the document into two parts (e.g., title and abstract, etc) which construct a coupled text pair. Then, we adopt negative sampling to construct uncoupled text pairs whose two parts are from different documents. Finally, we train the model to judge whether the text pair is coupled or uncoupled and use the obtained embedding of coupled text pairs as the embedding of documents. We perform experiments on three datasets for one information retrieval task and two recommendation tasks. The experimental results verify the effectiveness of the proposed CTPE model.

cs.CL↗

arXiv2021

JAS-GAN: Generative Adversarial Network Based Joint Atrium and Scar Segmentations on Unbalanced Atrial Targets

Automated and accurate segmentations of left atrium (LA) and atrial scars from late gadolinium-enhanced cardiac magnetic resonance (LGE CMR) images are in high demand for quantifying atrial scars. The previous quantification of atrial scars relies on a two-phase segmentation for LA and atrial scars due to their large volume difference (unbalanced atrial targets). In this paper, we propose an inter-cascade generative adversarial network, namely JAS-GAN, to segment the unbalanced atrial targets from LGE CMR images automatically and accurately in an end-to-end way. Firstly, JAS-GAN investigates an adaptive attention cascade to automatically correlate the segmentation tasks of the unbalanced atrial targets. The adaptive attention cascade mainly models the inclusion relationship of the two unbalanced atrial targets, where the estimated LA acts as the attention map to adaptively focus on the small atrial scars roughly. Then, an adversarial regularization is applied to the segmentation tasks of the unbalanced atrial targets for making a consistent optimization. It mainly forces the estimated joint distribution of LA and atrial scars to match the real ones. We evaluated the performance of our JAS-GAN on a 3D LGE CMR dataset with 192 scans. Compared with the state-of-the-art methods, our proposed approach yielded better segmentation performance (Average Dice Similarity Coefficient (DSC) values of 0.946 and 0.821 for LA and atrial scars, respectively), which indicated the effectiveness of our proposed approach for segmenting unbalanced atrial targets.

eess.IV↗

arXiv2021

Rescuing Deep Hashing from Dead Bits Problem

Deep hashing methods have shown great retrieval accuracy and efficiency in large-scale image retrieval. How to optimize discrete hash bits is always the focus in deep hashing methods. A common strategy in these methods is to adopt an activation function, e.g. $\operatorname{sigmoid}(\cdot)$ or $\operatorname{tanh}(\cdot)$, and minimize a quantization loss to approximate discrete values. However, this paradigm may make more and more hash bits stuck into the wrong saturated area of the activation functions and never escaped. We call this problem "Dead Bits Problem~(DBP)". Besides, the existing quantization loss will aggravate DBP as well. In this paper, we propose a simple but effective gradient amplifier which acts before activation functions to alleviate DBP. Moreover, we devise an error-aware quantization loss to further alleviate DBP. It avoids the negative effect of quantization loss based on the similarity between two images. The proposed gradient amplifier and error-aware quantization loss are compatible with a variety of deep hashing methods. Experimental results on three datasets demonstrate the efficiency of the proposed gradient amplifier and the error-aware quantization loss.

cs.CV↗