Search

Text-to-SQL systems, which convert natural language queries into SQL commands, have seen significant progress primarily for the SQLite dialect. However, adapting these systems to other SQL dialects like BigQuery and PostgreSQL remains a challenge due to the diversity in SQL syntax and functions. We introduce SQL-GEN, a framework for generating high-quality dialect-specific synthetic data guided by dialect-specific tutorials, and demonstrate its effectiveness in creating training datasets for multiple dialects. Our approach significantly improves performance, by up to 20\%, over previous methods and reduces the gap with large-scale human-annotated datasets. Moreover, combining our synthetic data with human-annotated data provides additional performance boosts of 3.3\% to 5.6\%. We also introduce a novel Mixture of Experts (MoE) initialization method that integrates dialect-specific models into a unified system by merging self-attention layers and initializing the gates with dialect-specific keywords, further enhancing performance across different SQL dialects.

The integration of artificial intelligence (AI) with radiology marks a transformative era in medicine. Vision foundation models have been adopted to enhance radiologic imaging analysis. However, the distinct complexities of radiologic 2D and 3D radiologic data pose unique challenges that existing models, pre-trained on general non-medical images, fail to address adequately. To bridge this gap and capitalize on the diagnostic precision required in radiologic imaging, we introduce Radiologic Contrastive Language-Image Pre-training (RadCLIP): a cross-modal vision-language foundational model that harnesses Vision Language Pre-training (VLP) framework to improve radiologic image analysis. Building upon Contrastive Language-Image Pre-training (CLIP), RadCLIP incorporates a slice pooling mechanism tailored for volumetric image analysis and is pre-trained using a large and diverse dataset of radiologic image-text pairs. The RadCLIP was pre-trained to effectively align radiologic images with their corresponding text annotations, creating a robust vision backbone for radiologic images. Extensive experiments demonstrate RadCLIP's superior performance in both uni-modal radiologic image classification and cross-modal image-text matching, highlighting its significant promise for improving diagnostic accuracy and efficiency in clinical settings. Our Key contributions include curating a large dataset with diverse radiologic 2D/3D radiologic image-text pairs, a slice pooling adapter using an attention mechanism for integrating 2D images, and comprehensive evaluations of RadCLIP on various radiologic downstream tasks.

The integration of different imaging modalities, such as structural, diffusion tensor, and functional magnetic resonance imaging, with deep learning models has yielded promising outcomes in discerning phenotypic characteristics and enhancing disease diagnosis. The development of such a technique hinges on the efficient fusion of heterogeneous multimodal features, which initially reside within distinct representation spaces. Naively fusing the multimodal features does not adequately capture the complementary information and could even produce redundancy. In this work, we present a novel joint self-supervised and supervised contrastive learning method to learn the robust latent feature representation from multimodal MRI data, allowing the projection of heterogeneous features into a shared common space, and thereby amalgamating both complementary and analogous information across various modalities and among similar subjects. We performed a comparative analysis between our proposed method and alternative deep multimodal learning approaches. Through extensive experiments on two independent datasets, the results demonstrated that our method is significantly superior to several other deep multimodal learning methods in predicting abnormal neurodevelopment. Our method has the capability to facilitate computer-aided diagnosis within clinical practice, harnessing the power of multimodal data., Comment: 35 pages. Submitted to journal

Dissipation is a common occurrence in real-world systems and is generally considered to be detrimental to transport. In this study, we examine the transport properties of a narrow quantum anomalous Hall system with dissipation applied on one edge. When the Fermi level resides within the hybridization gap, we find that while transport is suppressed on one edge, it is significantly enhanced on the other. We reveal that this enhancement arises from dissipation-induced gap closure, which is deeply rooted in the point gap topology of the system, resulting in a reduction of the decaying coefficient. When the dissipation is very large, we find that the low-energy physics is nearly indistinguishable from a narrower system, whose dissipation amplitude is inversely proportional to that of the original one. To get more physical intuition, we demonstrate that the low-energy physics can be well captured by a pair of coupled counter-propagating chiral edge states, one of which has a modified group velocity and an effective dissipation. We also briefly discuss the possible experimental realizations of this enhanced unidirectional transport.

Non-Hermitian systems have garnered significant attention due to the emergence of novel topology of complex spectra and skin modes. However, investigating transport phenomena in such systems faces obstacles stemming from the non-unitary nature of time evolution. Here, we establish the continuity equation for a general non-Hermitian Hamiltonian in the Schr\"odinger picture. It attributes the universal non-conservativity to the anti-commutation relationship between particle number and non-Hermitian terms. Our work derives a comprehensive current formula for non-Hermitian systems using Green's function, applicable to both time-dependent and steady-state responses. To demonstrate the validity of our approach, we calculate the local current in models with one-dimensional and two-dimensional settings, incorporating scattering potentials. The spatial distribution of local current highlights the widespread non-Hermitian phenomena, including skin modes, non-reciprocal quantum dots, and corner states. Our findings offer valuable insights for advancing theoretical and experimental research in the transport of non-Hermitian systems.

Multi-view clustering (MVC) has had significant implications in cross-modal representation learning and data-driven decision-making in recent years. It accomplishes this by leveraging the consistency and complementary information among multiple views to cluster samples into distinct groups. However, as contrastive learning continues to evolve within the field of computer vision, self-supervised learning has also made substantial research progress and is progressively becoming dominant in MVC methods. It guides the clustering process by designing proxy tasks to mine the representation of image and video data itself as supervisory information. Despite the rapid development of self-supervised MVC, there has yet to be a comprehensive survey to analyze and summarize the current state of research progress. Therefore, this paper explores the reasons and advantages of the emergence of self-supervised MVC and discusses the internal connections and classifications of common datasets, data issues, representation learning methods, and self-supervised learning methods. This paper does not only introduce the mechanisms for each category of methods but also gives a few examples of how these techniques are used. In the end, some open problems are pointed out for further investigation and development.

Studying protein interactions at low temperatures has important implications for optimizing cryostorage processes of biological tissue, food, and protein-based drugs. One of the major challenges is related to the formation of ice nanocrystals which can occur even in the presence of cryoprotectants and can lead to protein denaturation. Here, using a combination of small- and wide-angle X-ray scattering (SAXS and WAXS), we investigate the structural evolution of concentrated Lysozyme solutions in a cryoprotected glycerol-water mixture upon cooling from room temperature (T=300 K) down to cryogenic temperatures (T=195 K). Upon cooling, we observe a transition near the melting temperature of the solution (T~245 K), which manifests both in the temperature dependence of the scattering intensity peak position reflecting protein-protein length scales (SAXS) and the interatomic distances within the solvent (WAXS). Upon thermal cycling, a hysteresis is observed in the scattering intensity, which is attributed to the formation of nanocrystallites in the order of 10 nm. The experimental data is well described by the two-Yukawa model, which indicates temperature-dependent changes in the short-range attraction of the protein-protein interaction potential. Our results demonstrate that the nanocrystal growth yields effectively stronger protein-protein attraction and influences the protein pair distribution function beyond the first coordination shell.

Energy dissipation is of fundamental interest and crucial importance in quantum systems. However, whether energy dissipation can emerge inside topological systems remains a question, especially when charge transport is topologically protected and quantized. As a hallmark, we propose a microscopic picture that illustrates energy dissipation in the quantum Hall (QH) plateau regime of graphene. Despite the quantization of Hall, longitudinal, and two-probe resistances (dubbed as the quantum limit), we find that the energy dissipation emerges in the form of Joule heat. By analyzing the energy distribution of electrons, it is found that electrons can evolve between equilibrium and non-equilibrium without inducing extra two-probe resistance. The relaxation of non-equilibrium electrons results in the dissipation of energy along the QH edge states. Eventually, we suggest probing the phenomenon by measuring local temperature increases in experiments and reconsidering the dissipation typically ignored in realistic topological circuits., Comment: 7 pages, 4 figures;The computer programs associated with this paper are available via the GitHub website (https: //github.com/meplum-li/dissipation_calc.git)

The computer-aided disease diagnosis from radiomic data is important in many medical applications. However, developing such a technique relies on annotating radiological images, which is a time-consuming, labor-intensive, and expensive process. In this work, we present the first novel collaborative self-supervised learning method to solve the challenge of insufficient labeled radiomic data, whose characteristics are different from text and image data. To achieve this, we present two collaborative pretext tasks that explore the latent pathological or biological relationships between regions of interest and the similarity and dissimilarity information between subjects. Our method collaboratively learns the robust latent feature representations from radiomic data in a self-supervised manner to reduce human annotation efforts, which benefits the disease diagnosis. We compared our proposed method with other state-of-the-art self-supervised learning methods on a simulation study and two independent datasets. Extensive experimental results demonstrated that our method outperforms other self-supervised learning methods on both classification and regression tasks. With further refinement, our method shows the potential advantage in automatic disease diagnosis with large-scale unlabeled data available., Comment: 14 pages, 7 figures

Recently, a type of Hall effect due to an unusual layer-locked Berry curvature called the layer Hall effect (LHE) has been reported in the even-layered two-dimensional antiferromagnetic (AFM) MnBi2Te4 [A. Gao et.al, Nature 595, 521 (2021)]. In this work, we report that the quantization of LHE, which we call the quantum anomalous layer Hall effect (QALHE), can be realized in MnBi2Te4. The QALHE originates from kicking a layer-locked Berry-curvature monopole out of the Fermi sea by a vertielectric cal field. Remarkably, we demonstrate that the electric-field reversal can switch the sign of the quantized Hall conductance of QALHE in the even-layered AFM phase. The QALHE can also be realized in the ferromagnetic phase. These results provide a promising way toward the electric engineering of the Berry curvature monopoles and quantized-layered transport in topological magnets., Comment: 6 pages, 4 figures (+ supplementary materials 6 pages, 4 figures)