Searcharxiv⌕ Search

arXiv subjects

Xin Gao

Publications and source records attributed to Xin Gao.

At least 217 records · Page 12Linked to original sources

Towards artificial general intelligence via a multimodal foundation model

The fundamental goal of artificial intelligence (AI) is to mimic the core cognitive activities of human. Despite tremendous success in the AI research, most of existing methods have only single-cognitive ability. To overcome this limitation and take a solid step towards artificial general intelligence (AGI), we develop a foundation model pre-trained with huge multimodal data, which can be quickly adapted for various downstream cognitive tasks. To achieve this goal, we propose to pre-train our foundation model by self-supervised learning with weak semantic correlation data crawled from the Internet and show that promising results can be obtained on a wide range of downstream tasks. Particularly, with the developed model-interpretability tools, we demonstrate that strong imagination ability is now possessed by our foundation model. We believe that our work makes a transformative stride towards AGI, from our common practice of "weak or narrow AI" to that of "strong or generalized AI".

cs.AI↗

Target-aware Abstractive Related Work Generation with Contrastive Learning

The related work section is an important component of a scientific paper, which highlights the contribution of the target paper in the context of the reference papers. Authors can save their time and effort by using the automatically generated related work section as a draft to complete the final related work. Most of the existing related work section generation methods rely on extracting off-the-shelf sentences to make a comparative discussion about the target work and the reference papers. However, such sentences need to be written in advance and are hard to obtain in practice. Hence, in this paper, we propose an abstractive target-aware related work generator (TAG), which can generate related work sections consisting of new sentences. Concretely, we first propose a target-aware graph encoder, which models the relationships between reference papers and the target paper with target-centered attention mechanisms. In the decoding process, we propose a hierarchical decoder that attends to the nodes of different levels in the graph with keyphrases as semantic indicators. Finally, to generate a more informative related work, we propose multi-level contrastive optimization objectives, which aim to maximize the mutual information between the generated related work with the references and minimize that with non-references. Extensive experiments on two public scholar datasets show that the proposed model brings substantial improvements over several strong baselines in terms of automatic and tailored human evaluations.

cs.CL↗

Context Attention Network for Skeleton Extraction

Skeleton extraction is a task focused on providing a simple representation of an object by extracting the skeleton from the given binary or RGB image. In recent years many attractive works in skeleton extraction have been made. But as far as we know, there is little research on how to utilize the context information in the binary shape of objects. In this paper, we propose an attention-based model called Context Attention Network (CANet), which integrates the context extraction module in a UNet architecture and can effectively improve the ability of network to extract the skeleton pixels. Meanwhile, we also use some novel techniques including distance transform, weight focal loss to achieve good results on the given dataset. Finally, without model ensemble and with only 80% of the training images, our method achieves 0.822 F1 score during the development phase and 0.8507 F1 score during the final phase of the Pixel SkelNetOn Competition, ranking 1st place on the leaderboard.

cs.CV↗

Modeling COVID-19 vaccine-induced immunological memory development and its links to antibody level and infectiousness

COVID-19 vaccines have proven to be effective against SARS-CoV-2 infection. However, the dynamics of vaccine-induced immunological memory development and neutralizing antibodies generation are not fully understood, limiting vaccine development and vaccination regimen determination. Herein, we constructed a mathematical model to characterize the vaccine-induced immune response based on fitting the viral infection and vaccination datasets. With the example of CoronaVac, we revealed the association between vaccine-induced immunological memory development and neutralizing antibody levels. The establishment of the intact immunological memory requires more than 6 months after the first and second doses, after that a booster shot can induce high levels neutralizing antibodies. By introducing the maximum viral load and recovery time after viral infection, we quantitatively studied the protective effect of vaccines against viral infection. Accordingly, we optimized the vaccination regimen, including dose and vaccination timing, and predicted the effect of the fourth dose. Last, by combining the viral transmission model, we showed the suppression of virus transmission by vaccination, which may be instructive for the development of public health policies.

q-bio.QM↗

Evolution of barchan dune interactions investigated by a downscaled water tunnel experiment: the temporal characteristics and a soliton-like behavior

This paper reports a downscaled water tunnel experiment to study the temporal characteristics of a double dune interaction system and the new pattern of dune interaction when the initial mass ratio of the two dunes is large. These topics are useful for a comprehensive understanding of the dune interaction system but were rarely covered before. The turnover time scale under dune interaction is defined, and its time averaged value is found to have a nonmonotonic relationship with the initial mass ratio. A nonmonotonic relationship is also found between the convexity of the downstream dune tip and the initial mass ratio. The stationary points of the two nonmonotonic curves above correspond to the same dune interaction pattern named 'exchange-chasing', which is considered indispensable in the classification map of dune interactions. The upstream dune acts as an energy transmitter between fluid flow and the downstream dune. A soliton-like behavior occurs when the downstream dune enlarges, where a small dune is detached from the downstream dune tip and gets passed by the upstream dune approximately without mass exchange. The activity of such temporary soliton is found to be negatively related with the initial dune spacing and positively related with the initial mass ratio.

physics.geo-ph↗

Orientifold Calabi-Yau Threefolds with Divisor Involutions and String Landscape

We establish an orientifold Calabi-Yau threefold database for $h^{1,1}(X) \leq 6$ by considering non-trivial $\mathbb{Z}_{2}$ divisor exchange involutions, using a toric Calabi-Yau database (http://www.rossealtman.com/toriccy/). We first determine the topology for each individual divisor (Hodge diamond), then identify and classify the proper involutions which are globally consistent across all disjoint phases of the Kähler cone for each unique geometry. Each of the proper involutions will result in an orientifold Calabi-Yau manifold. Then we clarify all possible fixed loci under the proper involution, thereby determining the locations of different types of $O$-planes. It is shown that under the proper involutions, one typically ends up with a system of $O3/O7$-planes, and most of these will further admit naive Type IIB string vacua.The geometries with freely acting involutions are also determined. We further determine the splitting of the Hodge numbers into odd/even parity in the orbifold limit. The final result is a class of orientifold Calabi-Yau threefolds with non-trivial odd class cohomology $h^{1,1}_{-}(X / σ^*) \neq 0$.

hep-th↗

Applying machine learning to the Calabi-Yau orientifolds with string vacua

We use the machine learning technique to search the polytope which can result in an orientifold Calabi-Yau hypersurface and the "naive Type IIB string vacua". We show that neural networks can be trained to give a high accuracy for classifying the orientifold property and vacua based on the newly generated orientifold Calabi-Yau database with $h^{1,1}(X) \leq 6$ arXiv:2111.03078. This indicates the orientifold symmetry may already be encoded in the polytope structure. In the end, we try to use the trained neural networks model to go beyond the database and predict the orientifold signal of polytope for higher $h^{1,1}(X)$.

hep-th↗

Reentrance of metal-insulator transition and magnetic competitions on a triangular lattice with second nearest-neighbor hopping

The $120^{\circ}$ antiferromagnetism (AFM) is widely believed as the magnetic ground state of the triangular systems because of the geometrical frustration. The emergence of novel magnetism, such as the row-wise AFM in Mn/Cu(111) and Sn/Si(111), reveals the importance of the longer-range hopping on magnetic competitions in realistic material systems. By utilizing advanced many-body techniques, we systematically studied the isotropic triangular Hubbard model with second nearest-neighbor hopping $t^{\prime}$, including both the single- and the two-particle responses. We found that both electronic and magnetic phase transitions show a clear dependence on $t^{\prime}/t$. Consequently, we observed a remarkable reentrance of the metal-insulator transition and a crossover between the $120^{\circ}$- and the row-wise AFM. The Fermi surface (FS) shows two distinct structures with the nesting vectors consistent with the magnetic correlations. When $t^{\prime}$ evolves from 0 to 1, the correlated Fermi surface demonstrates a Lifschitz transition between the two nesting structures, and exotic phases like the featureless insulating state can be realized. Our work sheds light on the engineering of electronic and magnetic correlations of correlated triangular surfaces via longer-range hopping. The rich phase diagram and the high degree of tunability make the triangular lattice with longer-range hopping a more realistic platform to study the emergent magnetic competitions.

cond-mat.str-el↗

Weakly Supervised High-Fidelity Clothing Model Generation

The development of online economics arouses the demand of generating images of models on product clothes, to display new clothes and promote sales. However, the expensive proprietary model images challenge the existing image virtual try-on methods in this scenario, as most of them need to be trained on considerable amounts of model images accompanied with paired clothes images. In this paper, we propose a cheap yet scalable weakly-supervised method called Deep Generative Projection (DGP) to address this specific scenario. Lying in the heart of the proposed method is to imitate the process of human predicting the wearing effect, which is an unsupervised imagination based on life experience rather than computation rules learned from supervisions. Here a pretrained StyleGAN is used to capture the practical experience of wearing. Experiments show that projecting the rough alignment of clothing and body onto the StyleGAN space can yield photo-realistic wearing results. Experiments on real scene proprietary model images demonstrate the superiority of DGP over several state-of-the-art supervised methods when generating clothing model images.

cs.CV↗

On Admissible Positions of Transonic Shocks for Steady Isothermal Euler Flows in a Horizontal Flat Nozzle under Vertical Gravity

In this paper we are concerned with the existence of transonic shocks for 2-D steady isothermal Euler flows in a horizontal flat nozzle under vertical gravity. In particular, we focus on the contribution of the vertical gravity in determining the position of the shock front. For steady horizontal flows, the existence of normal shocks with the position of the shock front being arbitrary in the nozzle can be easily established. This paper will try to determine the position of the shock front as the state of the flow at the entrance of the nozzle and the pressure at the exit are slightly perturbed. To this end, this paper proposes a free boundary problem of the linearized Euler system with vertical gravity, whose solution could be an initial approximation for the shock solution with the free boundary being the approximation for the shock front. Due to the existence of the vertical gravity, difficulties arise in solving the boundary value problem in the approximate subsonic domain behind the shock front. The linearized Euler system is elliptic-hyperbolic composite for subsonic flows, and the elliptic part and the hyperbolic part are coupled in the $0$-order terms depending on the acceleration of gravity $g$. Moreover, the coefficients are not constants since the unperturbed shock solution depends on the vertical variable. New ideas and techniques are developed to deal with these difficulties and, under certain sufficient conditions on the perturbation, the existence of the solution to the proposed free boundary problem is established as the acceleration of gravity $g>0$ and the perturbation are sufficiently small. Then, with the obtained initial approximation of the shock solution, a nonlinear iteration scheme can be constructed which leads to a transonic shock solution with the position of the shock front being close to the initial approximating position.

math.AP↗

Protein-RNA interaction prediction with deep learning: Structure matters

Protein-RNA interactions are of vital importance to a variety of cellular activities. Both experimental and computational techniques have been developed to study the interactions. Due to the limitation of the previous database, especially the lack of protein structure data, most of the existing computational methods rely heavily on the sequence data, with only a small portion of the methods utilizing the structural information. Recently, AlphaFold has revolutionized the entire protein and biology field. Foreseeably, the protein-RNA interaction prediction will also be promoted significantly in the upcoming years. In this work, we give a thorough review of this field, surveying both the binding site and binding preference prediction problems and covering the commonly used datasets, features, and models. We also point out the potential challenges and opportunities in this field. This survey summarizes the development of the RBP-RNA interaction field in the past and foresees its future development in the post-AlphaFold era.

q-bio.BM↗

A Data-Centric Framework for Composable NLP Workflows

Empirical natural language processing (NLP) systems in application domains (e.g., healthcare, finance, education) involve interoperation among multiple components, ranging from data ingestion, human annotation, to text retrieval, analysis, generation, and visualization. We establish a unified open-source framework to support fast development of such sophisticated NLP workflows in a composable manner. The framework introduces a uniform data representation to encode heterogeneous results by a wide range of NLP tasks. It offers a large repository of processors for NLP tasks, visualization, and annotation, which can be easily assembled with full interoperability under the unified representation. The highly extensible framework allows plugging in custom processors from external off-the-shelf NLP and deep learning libraries. The whole framework is delivered through two modularized yet integratable open-source projects, namely Forte (for workflow infrastructure and NLP function processors) and Stave (for user interaction, visualization, and annotation).

cs.CL↗

Shape Controllable Virtual Try-on for Underwear Models

Image virtual try-on task has abundant applications and has become a hot research topic recently. Existing 2D image-based virtual try-on methods aim to transfer a target clothing image onto a reference person, which has two main disadvantages: cannot control the size and length precisely; unable to accurately estimate the user's figure in the case of users wearing thick clothes, resulting in inaccurate dressing effect. In this paper, we put forward an akin task that aims to dress clothing for underwear models. %, which is also an urgent need in e-commerce scenarios. To solve the above drawbacks, we propose a Shape Controllable Virtual Try-On Network (SC-VTON), where a graph attention network integrates the information of model and clothing to generate the warped clothing image. In addition, the control points are incorporated into SC-VTON for the desired clothing shape. Furthermore, by adding a Splitting Network and a Synthesis Network, we can use clothing/model pair data to help optimize the deformation module and generalize the task to the typical virtual try-on task. Extensive experiments show that the proposed method can achieve accurate shape control. Meanwhile, compared with other methods, our method can generate high-resolution results with detailed textures.

cs.CV↗

Double Similarity Distillation for Semantic Image Segmentation

The balance between high accuracy and high speed has always been a challenging task in semantic image segmentation. Compact segmentation networks are more widely used in the case of limited resources, while their performances are constrained. In this paper, motivated by the residual learning and global aggregation, we propose a simple yet general and effective knowledge distillation framework called double similarity distillation (DSD) to improve the classification accuracy of all existing compact networks by capturing the similarity knowledge in pixel and category dimensions, respectively. Specifically, we propose a pixel-wise similarity distillation (PSD) module that utilizes residual attention maps to capture more detailed spatial dependencies across multiple layers. Compared with exiting methods, the PSD module greatly reduces the amount of calculation and is easy to expand. Furthermore, considering the differences in characteristics between semantic segmentation task and other computer vision tasks, we propose a category-wise similarity distillation (CSD) module, which can help the compact segmentation network strengthen the global category correlation by constructing the correlation matrix. Combining these two modules, DSD framework has no extra parameters and only a minimal increase in FLOPs. Extensive experiments on four challenging datasets, including Cityscapes, CamVid, ADE20K, and Pascal VOC 2012, show that DSD outperforms current state-of-the-art methods, proving its effectiveness and generality. The code and models will be publicly available.

cs.CV↗

2nd Place Solution for Waymo Open Dataset Challenge -- Real-time 2D Object Detection

In an autonomous driving system, it is essential to recognize vehicles, pedestrians and cyclists from images. Besides the high accuracy of the prediction, the requirement of real-time running brings new challenges for convolutional network models. In this report, we introduce a real-time method to detect the 2D objects from images. We aggregate several popular one-stage object detectors and train the models of variety input strategies independently, to yield better performance for accurate multi-scale detection of each category, especially for small objects. For model acceleration, we leverage TensorRT to optimize the inference time of our detection pipeline. As shown in the leaderboard, our proposed detection framework ranks the 2nd place with 75.00% L1 mAP and 69.72% L2 mAP in the real-time 2D detection track of the Waymo Open Dataset Challenges, while our framework achieves the latency of 45.8ms/frame on an Nvidia Tesla V100 GPU.

cs.CV↗

Asymmetric Loss Functions for Learning with Noisy Labels

Robust loss functions are essential for training deep neural networks with better generalization power in the presence of noisy labels. Symmetric loss functions are confirmed to be robust to label noise. However, the symmetric condition is overly restrictive. In this work, we propose a new class of loss functions, namely \textit{asymmetric loss functions}, which are robust to learning with noisy labels for various types of noise. We investigate general theoretical properties of asymmetric loss functions, including classification calibration, excess risk bound, and noise tolerance. Meanwhile, we introduce the asymmetry ratio to measure the asymmetry of a loss function. The empirical results show that a higher ratio would provide better noise tolerance. Moreover, we modify several commonly-used loss functions and establish the necessary and sufficient conditions for them to be asymmetric. Experimental results on benchmark datasets demonstrate that asymmetric loss functions can outperform state-of-the-art methods. The code is available at \href{https://github.com/hitcszx/ALFs}{https://github.com/hitcszx/ALFs}

cs.LG↗

The Reconfiguration Pattern of Individual Brain Metabolic Connectome for Parkinson's Disease Identification

Background: Positron Emission Tomography (PET) with 18F-fluorodeoxyglucose (18F-FDG) reveals metabolic abnormalities in Parkinson's disease (PD) at a systemic level. Previous metabolic connectome studies derived from groups of patients have failed to identify the individual neurophysiological details. We aim to establish an individual metabolic connectome method to characterize the aberrant connectivity patterns and topological alterations of the individual-level brain metabolic connectome and their diagnostic value in PD. Methods: The 18F-FDG PET data of 49 PD patients and 49 healthy controls (HCs) were recruited. Each individual's metabolic brain network was ascertained using the proposed Jensen-Shannon Divergence Similarity Estimation (JSSE) method. The intergroup difference of the individual's metabolic brain network and its global and local graph metrics were analyzed to investigate the metabolic connectome's alterations. The identification of the PD from HC individuals was used by the multiple kernel support vector machine (MK-SVM) to combine the information from connection and topological metrics. The validation was conducted using the nest leave-one-out cross-validation strategy to confirm the performance of the methods. Results: The proposed JSSE metabolic connectome method showed the most involved metabolic motor networks were PUT-PCG, THA-PCG, and SMA pathways in PD, which was similar to the typical group-level method, and yielded another detailed individual pathological connectivity in ACG-PCL, DCG-PHG and ACG pathways. These aberrant functional network measures exhibited an ideal classification performance in the identifying of PD individuals from HC individuals at an accuracy of up to 91.84%.

q-bio.NC↗

History-Aware Online Cache Placement in Fog-Assisted IoT Systems: An Integration of Learning and Control

In Fog-assisted IoT systems, it is a common practice to cache popular content at the network edge to achieve high quality of service. Due to uncertainties in practice such as unknown file popularities, cache placement scheme design is still an open problem with unresolved challenges: 1) how to maintain time-averaged storage costs under budgets, 2) how to incorporate online learning to aid cache placement to minimize performance loss (a.k.a. regret), and 3) how to exploit offline historical information to further reduce regret. In this paper, we formulate the cache placement problem with unknown file popularities as a constrained combinatorial multi-armed bandit (CMAB) problem. To solve the problem, we employ virtual queue techniques to manage time-averaged storage cost constraints, and adopt history-aware bandit learning methods to integrate offline historical information into the online learning procedure to handle the exploration-exploitation tradeoff. With an effective combination of online control and history-aware online learning, we devise a Cache Placement scheme with History-aware Bandit Learning called CPHBL. Our theoretical analysis and simulations show that CPHBL achieves a sublinear time-averaged regret bound. Moreover, the simulation results verify CPHBL's advantage over the deep reinforcement learning based approach.

cs.NI↗