arXiv subjects

Xin Xing

Publications and source records attributed to Xin Xing.

At least 19 recordsLinked to original sources

arXiv2026

Multiclass Calibration Assessment and Recalibration of Probability Predictions via the Linear Log Odds Calibration Function

Machine-generated probability predictions are essential in modern classification tasks such as image classification. A model is well calibrated when its predicted probabilities correspond to observed event frequencies. Despite the need for multicategory recalibration methods, existing methods are limited to (i) comparing calibration between two or more models rather than directly assessing the calibration of a single model, (ii) requiring under-the-hood model access, e.g., accessing logit-scale predictions within the layers of a neural network, and (iii) providing output which is difficult for human analysts to understand. To overcome (i)-(iii), we propose Multicategory Linear Log Odds (MCLLO) recalibration, which (i) includes a likelihood ratio hypothesis test to assess calibration, (ii) does not require under-the-hood access to models and is thus applicable on a wide range of classification problems, and (iii) can be easily interpreted. We demonstrate the effectiveness of the MCLLO method through simulations and three real-world case studies involving image classification via convolutional neural network, obesity analysis via random forest, and ecology via regression modeling. We compare MCLLO to four comparator recalibration techniques utilizing both our hypothesis test and the existing calibration metric Expected Calibration Error to show that our method works well alone and in concert with other methods.

stat.ML↗

arXiv2025

A L-infinity Norm Synthetic Control Approach

This paper reinterprets the Synthetic Control (SC) framework through the lens of weighting philosophy, arguing that the contrast between traditional SC and Difference-in-Differences (DID) reflects two distinct modeling mindsets: sparse versus dense weighting schemes. Rather than viewing sparsity as inherently superior, we treat it as a modeling choice simple but potentially fragile. We propose an L-infinity-regularized SC method that combines the strengths of both approaches. Like DID, it employs a denser weighting scheme that distributes weights more evenly across control units, enhancing robustness and reducing overreliance on a few control units. Like traditional SC, it remains flexible and data-driven, increasing the likelihood of satisfying the parallel trends assumption while preserving interpretability. We develop an interior point algorithm for efficient computation, derive asymptotic theory under weak dependence, and demonstrate strong finite-sample performance through simulations and real-world applications.

stat.ME↗

arXiv2025

MAE-SAM2: Mask Autoencoder-Enhanced SAM2 for Clinical Retinal Vascular Leakage Segmentation

We propose MAE-SAM2, a novel foundation model for retinal vascular leakage segmentation on fluorescein angiography images. Due to the small size and dense distribution of the leakage areas, along with the limited availability of labeled clinical data, this presents a significant challenge for segmentation tasks. Our approach integrates a Self-Supervised learning (SSL) strategy, Masked Autoencoder (MAE), with SAM2. In our implementation, we explore different loss functions and conclude a task-specific combined loss. Extensive experiments and ablation studies demonstrate that MAE-SAM2 outperforms several state-of-the-art models, achieving the highest Dice score and Intersection-over-Union (IoU). Compared to the original SAM2, our model achieves a $5\%$ performance improvement, highlighting the promise of foundation models with self-supervised pretraining in clinical imaging tasks.

q-bio.TO↗

arXiv2025

Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges

Due to the remarkable capabilities and growing impact of large language models (LLMs), they have been deeply integrated into many aspects of society. Thus, ensuring their alignment with human values and intentions has emerged as a critical challenge. This survey provides a comprehensive overview of practical alignment techniques, training protocols, and empirical findings in LLM alignment. We analyze the development of alignment methods across diverse paradigms, characterizing the fundamental trade-offs between core alignment objectives. Our analysis shows that while supervised fine-tuning enables basic instruction-following, preference-based methods offer more flexibility for aligning with nuanced human intent. We discuss state-of-the-art techniques, including Direct Preference Optimization (DPO), Constitutional AI, brain-inspired methods, and alignment uncertainty quantification (AUQ), highlighting their approaches to balancing quality and efficiency. We review existing evaluation frameworks and benchmarking datasets, emphasizing limitations such as reward misspecification, distributional robustness, and scalable oversight. We summarize strategies adopted by leading AI labs to illustrate the current state of practice. We conclude by outlining open problems in oversight, value pluralism, robustness, and continuous alignment. This survey aims to inform both researchers and practitioners navigating the evolving landscape of LLM alignment.

cs.AI↗

arXiv2025

Optimized auxiliary functions for robust mitigation of finite-size errors in periodic hybrid density functional theory

When calculating properties of periodic systems at the thermodynamic limit (TDL), the dominant source of finite size error (FSE) arises from the long-range Coulomb interaction, and can manifest as a slowly converging quadrature error when approximating an integral in the reciprocal space by a finite sum. The singularity subtraction (SS) method offers a systematic approach for reducing this quadrature error and thus the FSE. In this work, we first investigate the performance of the SS method in the simplest setting, aiming at reducing the FSE in exact exchange calculations by subtracting the Coulomb contribution with a single, adjustable Gaussian auxiliary function. We demonstrate that a simple fitting method can robustly estimate the optimal Gaussian width and leads to rapid convergence toward the TDL. Furthermore, we suggest new forms of the auxiliary function, whose optimal parameters could also be determined through least-squares fitting. For a range of semiconductors and insulators, the proposed auxiliary functions achieve robust, millihartree-level accuracy in hybrid density functional theory calculations, including cases with sparse k-meshes and large basis sets.

physics.comp-ph↗

arXiv2024

Mixed-View Panorama Synthesis using Geospatially Guided Diffusion

We introduce the task of mixed-view panorama synthesis, where the goal is to synthesize a novel panorama given a small set of input panoramas and a satellite image of the area. This contrasts with previous work which only uses input panoramas (same-view synthesis), or an input satellite image (cross-view synthesis). We argue that the mixed-view setting is the most natural to support panorama synthesis for arbitrary locations worldwide. A critical challenge is that the spatial coverage of panoramas is uneven, with few panoramas available in many regions of the world. We introduce an approach that utilizes diffusion-based modeling and an attention-based architecture for extracting information from all available input imagery. Experimental results demonstrate the effectiveness of our proposed method. In particular, our model can handle scenarios when the available panoramas are sparse or far from the location of the panorama we are attempting to synthesize. The project page is available at https://mixed-view.github.io

cs.CV↗

arXiv2024

Inverse Volume Scaling of Finite-Size Error in Periodic Coupled Cluster Theory

Coupled cluster theory is one of the most popular post-Hartree-Fock methods for ab initio molecular quantum chemistry. The finite-size error of the correlation energy in periodic coupled cluster calculations for three-dimensional insulating systems has been observed to satisfy the inverse volume scaling, even in the absence of any correction schemes. This is surprising, as simpler theories that utilize only a subset of the coupled cluster diagrams exhibit much slower decay of the finite-size error, which scales inversely with the length of the system. In this study, we review the current understanding of finite-size error in quantum chemistry methods for periodic systems. We introduce new tools that elucidate the mechanisms behind this phenomenon in the context of coupled cluster doubles calculations. This reconciles some seemingly paradoxical statements related to finite-size scaling. Our findings also show that singularity subtraction can be a powerful method to effectively reduce finite-size errors in practical quantum chemistry calculations for periodic systems.

math.NA↗

arXiv2024

The Staggered Mesh Method: Accurate Exact Exchange towards the Thermodynamic Limit for Solids

In periodic systems, the Hartree-Fock (HF) exchange energy exhibits the slowest convergence of all HF energy components as the system size approaches the thermodynamic limit. We demonstrate that the recently proposed staggered mesh method for Fock exchange energy [Xing, Li, and Lin, Math. Comp., 2024], which is specifically designed to sidestep certain singularities in exchange energy evaluation, can expedite the finite-size convergence rate for the exact exchange energy across a range of insulators and semiconductors when compared to the regular and truncated Coulomb methods. This remains true even for two computationally cheaper versions of this new method, which we call Non-SCF and Split-SCF staggered mesh. Additionally, a sequence of numerical tests on simple solids showcases the staggered mesh method's ability to improve convergence towards the thermodynamic limit for band gaps, bulk moduli, equilibrium lattice dimensions, energies, and phonon force constants.

physics.comp-ph↗

arXiv2023

LD-SDM: Language-Driven Hierarchical Species Distribution Modeling

We focus on species distribution modeling using global-scale presence-only data, leveraging geographical and environmental features to map species ranges, as in previous studies. However, we innovate by integrating taxonomic classification into our approach. Specifically, we propose using a large language model to extract a latent representation of the taxonomic classification from a textual prompt. This allows us to map the range of any taxonomic rank, including unseen species, without additional supervision. We also present a new proximity-aware evaluation metric, suitable for evaluating species distribution models, which addresses critical shortcomings of traditional metrics. We evaluated our model for species range prediction, zero-shot prediction, and geo-feature regression and found that it outperforms several state-of-the-art models.

cs.CV↗

arXiv2023

Unified analysis of finite-size error for periodic Hartree-Fock and second order Møller-Plesset perturbation theory

Despite decades of practice, finite-size errors in many widely used electronic structure theories for periodic systems remain poorly understood. For periodic systems using a general Monkhorst-Pack grid, there has been no comprehensive and rigorous analysis of the finite-size error in the Hartree-Fock theory (HF) and the second order Møller-Plesset perturbation theory (MP2), which are the simplest wavefunction based method, and the simplest post-Hartree-Fock method, respectively. Such calculations can be viewed as a multi-dimensional integral discretized with certain trapezoidal rules. Due to the Coulomb singularity, the integrand has many points of discontinuity in general, and standard error analysis based on the Euler-Maclaurin formula gives overly pessimistic results. The lack of analytic understanding of finite-size errors also impedes the development of effective finite-size correction schemes. We propose a unified analysis to obtain sharp convergence rates of finite-size errors for the periodic HF and MP2 theories. Our main technical advancement is a generalization of the result of [Lyness, 1976] for obtaining sharp convergence rates of the trapezoidal rule for a class of non-smooth integrands. Our result is applicable to three-dimensional bulk systems as well as low dimensional systems (such as nanowires and 2D materials). Our unified analysis also allows us to prove the effectiveness of the Madelung-constant correction to the Fock exchange energy, and the effectiveness of a recently proposed staggered mesh method for periodic MP2 calculations [Xing, Li, Lin, J. Chem. Theory Comput. 2021]. Our analysis connects the effectiveness of the staggered mesh method with integrands with removable singularities, and suggests a new staggered mesh method for reducing finite-size errors of periodic HF calculations.

physics.comp-ph↗

arXiv2023

Vision-Language Pseudo-Labels for Single-Positive Multi-Label Learning

This paper presents a novel approach to Single-Positive Multi-label Learning. In general multi-label learning, a model learns to predict multiple labels or categories for a single input image. This is in contrast with standard multi-class image classification, where the task is predicting a single label from many possible labels for an image. Single-Positive Multi-label Learning (SPML) specifically considers learning to predict multiple labels when there is only a single annotation per image in the training data. Multi-label learning is in many ways a more realistic task than single-label learning as real-world data often involves instances belonging to multiple categories simultaneously; however, most common computer vision datasets predominantly contain single labels due to the inherent complexity and cost of collecting multiple high quality annotations for each instance. We propose a novel approach called Vision-Language Pseudo-Labeling (VLPL), which uses a vision-language model to suggest strong positive and negative pseudo-labels, and outperforms the current SOTA methods by 5.5% on Pascal VOC, 18.4% on MS-COCO, 15.2% on NUS-WIDE, and 8.4% on CUB-Birds. Our code and data are available at https://github.com/mvrl/VLPL.

cs.CV↗

arXiv2023

On the testing of multiple hypothesis in sliced inverse regression

We consider the multiple testing of the general regression framework aiming at studying the relationship between a univariate response and a p-dimensional predictor. To test the hypothesis of the effect of each predictor, we construct an Angular Balanced Statistic (ABS) based on the estimator of the sliced inverse regression without assuming a model of the conditional distribution of the response. According to the developed limiting distribution results in this paper, we have shown that ABS is asymptotically symmetric with respect to zero under the null hypothesis. We then propose a Model-free multiple Testing procedure using Angular balanced statistics (MTA) and show theoretically that the false discovery rate of this method is less than or equal to a designated level asymptotically. Numerical evidence has shown that the MTA method is much more powerful than its alternatives, subject to the control of the false discovery rate.

stat.ME↗

arXiv2023

Finite-size effects in periodic coupled cluster calculations

We provide the first rigorous study of the finite-size error in the simplest and representative coupled cluster theory, namely the coupled cluster doubles (CCD) theory, for gapped periodic systems. Assuming that the CCD equations are solved using exact Hartree-Fock orbitals and orbital energies, we prove that the convergence rate of finite-size error scales as $\mathscr{O}(N_\mathbf{k}^{-\frac13})$, where $N_{\mathbf{k}}$ is the number of discretization point in the Brillouin zone and characterizes the system size. Our analysis shows that the dominant error lies in the coupled cluster amplitude calculation, and the convergence of the finite-size error in energy calculations can be boosted to $\mathscr{O}(N_\mathbf{k}^{-1})$ with accurate amplitudes. This also provides the first proof of the scaling of the finite-size error in the third order Møller-Plesset perturbation theory (MP3) for periodic systems.

math.NA↗

arXiv2022

Robust Flow-based Conformal Inference (FCI) with Statistical Guarantee

Conformal prediction aims to determine precise levels of confidence in predictions for new objects using past experience. However, the commonly used exchangeable assumptions between the training data and testing data limit its usage in dealing with contaminated testing sets. In this paper, we develop a novel flow-based conformal inference (FCI) method to build predictive sets and infer outliers for complex and high-dimensional data. We leverage ideas from adversarial flow to transfer the input data to a random vector with known distributions. Our roundtrip transformation can map the input data to a low-dimensional space, meanwhile reserving the conditional distribution of input data given each class label, which enables us to construct a non-conformity score for uncertainty quantification. Our approach is applicable and robust when the testing data is contaminated. We evaluate our method, robust flow-based conformal inference, on benchmark datasets. We find that it produces effective predictive sets and accurate outlier detection and is more powerful relative to competing approaches.

stat.ML↗

arXiv2022

Causality for Inherently Explainable Transformers: CAT-XPLAIN

There have been several post-hoc explanation approaches developed to explain pre-trained black-box neural networks. However, there is still a gap in research efforts toward designing neural networks that are inherently explainable. In this paper, we utilize a recently proposed instance-wise post-hoc causal explanation method to make an existing transformer architecture inherently explainable. Once trained, our model provides an explanation in the form of top-$k$ regions in the input space of the given instance contributing to its decision. We evaluate our method on binary classification tasks using three image datasets: MNIST, FMNIST, and CIFAR. Our results demonstrate that compared to the causality-based post-hoc explainer model, our inherently explainable model achieves better explainability results while eliminating the need of training a separate explainer model. Our code is available at https://github.com/mvrl/CAT-XPLAIN.

cs.CV↗

arXiv2022

Staggered mesh method for correlation energy calculations of solids: Random phase approximation in direct ring coupled cluster doubles and adiabatic connection formalisms

We propose a staggered mesh method for correlation energy calculations of periodic systems under the random phase approximation (RPA), which generalizes the recently developed staggered mesh method for periodic second order Møller-Plesset perturbation theory (MP2) calculations [Xing, Li, Lin, JCTC 2021]. Compared to standard RPA calculations, the staggered mesh method introduces negligible additional computational cost. It avoids a significant portion of the finite-size error, and can be asymptotically advantageous for quasi-1D systems and certain quasi-2D and 3D systems with high symmetries. We demonstrate the applicability of the method using two different formalisms: the direct ring coupled cluster doubles (drCCD) theory, and the adiabatic-connection (AC) fluctuation-dissipation theory. In the drCCD formalism, the second order screened exchange (SOSEX) correction can also be readily obtained using the staggered mesh method. In the AC formalism, the staggered mesh method naturally avoids the need of performing "head/wing" corrections to the dielectric operator. The effectiveness of the staggered mesh method for insulating systems is theoretically justified by investigating the finite-size error of each individual perturbative term in the RPA correlation energy, expanded as an infinite series of terms associated with ring diagrams. As a side contribution, our analysis provides a proof that the finite-size error of each perturbative term of standard RPA and SOSEX calculations scales as $\mathcal{O}(N_{\mathbf{k}}^{-1})$, where $N_{\mathbf{k}}$ is the number of grid points in a Monkhorst-Pack mesh.

physics.comp-ph↗

arXiv2021

Model-based Sparse Coding beyond Gaussian Independent Model

Sparse coding aims to model data vectors as sparse linear combinations of basis elements, but a majority of related studies are restricted to continuous data without spatial or temporal structure. A new model-based sparse coding (MSC) method is proposed to provide an effective and flexible framework for learning features from different data types: continuous, discrete, or categorical, and modeling different types of correlations: spatial or temporal. The specification of the sparsity level and how to adapt the estimation method to large-scale studies are also addressed. A fast EM algorithm is proposed for estimation, and its superior performance is demonstrated in simulation and multiple real applications such as image denoising, brain connectivity study, and spatial transcriptomic imaging.

stat.ME↗

arXiv2021

Staggered mesh method for correlation energy calculations of solids: Second order Møller-Plesset perturbation theory

The calculation of the MP2 correlation energy for extended systems can be viewed as a multi-dimensional integral in the thermodynamic limit, and the standard method for evaluating the MP2 energy can be viewed as a trapezoidal quadrature scheme. We demonstrate that existing analysis neglects certain contributions due to the non-smoothness of the integrand, and may significantly underestimate finite-size errors. We propose a new staggered mesh method, which uses two staggered Monkhorst-Pack meshes for occupied and virtual orbitals, respectively, to compute the MP2 energy. The staggered mesh method circumvents a significant error source in the standard method, in which certain quadrature nodes are always placed on points where the integrand is discontinuous. One significant advantage of the proposed method is that there are no tunable parameters, and the additional numerical effort needed can be negligible compared to the standard MP2 calculation. Numerical results indicate that the staggered mesh method can be particularly advantageous for quasi-1D systems, as well as quasi-2D and 3D systems with certain symmetries.

physics.comp-ph↗