Search arXiv⌕ Search

arXiv subjects

Xin Huang

Publications and source records attributed to Xin Huang.

At least 127 records · Page 7Linked to original sources
arXiv2024

Neural LightRig: Unlocking Accurate Object Normal and Material Estimation with Multi-Light Diffusion

Recovering the geometry and materials of objects from a single image is challenging due to its under-constrained nature. In this paper, we present Neural LightRig, a novel framework that boosts intrinsic estimation by leveraging auxiliary multi-lighting conditions from 2D diffusion priors. Specifically, 1) we first leverage illumination priors from large-scale diffusion models to build our multi-light diffusion model on a synthetic relighting dataset with dedicated designs. This diffusion model generates multiple consistent images, each illuminated by point light sources in different directions. 2) By using these varied lighting images to reduce estimation uncertainty, we train a large G-buffer model with a U-Net backbone to accurately predict surface normals and materials. Extensive experiments validate that our approach significantly outperforms state-of-the-art methods, enabling accurate surface normal and PBR material estimation with vivid relighting effects. Code and dataset are available on our project page at https://projects.zxhezexin.com/neural-lightrig.

cs.CV↗
arXiv2024

Style3D: Attention-guided Multi-view Style Transfer for 3D Object Generation

We present Style3D, a novel approach for generating stylized 3D objects from a content image and a style image. Unlike most previous methods that require case- or style-specific training, Style3D supports instant 3D object stylization. Our key insight is that 3D object stylization can be decomposed into two interconnected processes: multi-view dual-feature alignment and sparse-view spatial reconstruction. We introduce MultiFusion Attention, an attention-guided technique to achieve multi-view stylization from the content-style pair. Specifically, the query features from the content image preserve geometric consistency across multiple views, while the key and value features from the style image are used to guide the stylistic transfer. This dual-feature alignment ensures that spatial coherence and stylistic fidelity are maintained across multi-view images. Finally, a large 3D reconstruction model is introduced to generate coherent stylized 3D objects. By establishing an interplay between structural and stylistic features across multiple views, our approach enables a holistic 3D stylization process. Extensive experiments demonstrate that Style3D offers a more flexible and scalable solution for generating style-consistent 3D assets, surpassing existing methods in both computational efficiency and visual quality.

cs.CV↗
arXiv2024

An Automated Data Mining Framework Using Autoencoders for Feature Extraction and Dimensionality Reduction

This study proposes an automated data mining framework based on autoencoders and experimentally verifies its effectiveness in feature extraction and data dimensionality reduction. Through the encoding-decoding structure, the autoencoder can capture the data's potential characteristics and achieve noise reduction and anomaly detection, providing an efficient and stable solution for the data mining process. The experiment compared the performance of the autoencoder with traditional dimensionality reduction methods (such as PCA, FA, T-SNE, and UMAP). The results showed that the autoencoder performed best in terms of reconstruction error and root mean square error and could better retain data structure and enhance the generalization ability of the model. The autoencoder-based framework not only reduces manual intervention but also significantly improves the automation of data processing. In the future, with the advancement of deep learning and big data technology, the autoencoder method combined with a generative adversarial network (GAN) or graph neural network (GNN) is expected to be more widely used in the fields of complex data processing, real-time data analysis and intelligent decision-making.

cs.LG↗
arXiv2024

LAMOST J171013+532646: a detached short-period non-eclipsing hot subdwarf + white dwarf binary

We present an analysis of LAMOST J171013.211+532646.04 (hereafter J1710), a binary system comprising a hot subdwarf B star (sdB) and a white dwarf (WD) companion. Multi-epoch spectroscopy reveals an orbital period of 109.20279 minutes, consistent with TESS and ZTF photometric data, marking it as the sixth detached system known to harbor a WD companion with a period less than two hours. J1710 is remarkably close to Earth, situated at a distance of only \(350.68^{+4.20}_{-4.21} \, \mathrm{pc}\), with a GAIA G-band magnitude of 12.59, rendering it conducive for continuous observations. The spectral temperature is around 25164 K, in agreement with SED fitting results (\(25301^{+839}_{-743} \, \mathrm{K}\)). The TESS light curve displays ellipsoidal variation and Doppler beaming without eclipsing features. Through fitting the TESS light curve using the Wilson-Devinney code, we determined the masses for the sdB (\(M_1 = 0.44^{+0.06}_{-0.07} \, M_{\odot}\)) and the compact object (\(M_2 = 0.54^{+0.10}_{-0.07} \, M_{\odot}\)), with the compact object likely being a WD. Furthermore, MESA models suggest that the sdB, with a helium core mass of 0.431 \(M_{\odot}\) and a hydrogen envelope mass of \(1.3 \times 10^{-3}\, M_{\odot}\), is in the early helium main-sequence phase. The MESA binary evolution shows that the J1710 system is expected to evolve into a double white dwarf system, making it an important source of low-frequency gravitational waves.

astro-ph.SR↗
arXiv2024

Material Anything: Generating Materials for Any 3D Object via Diffusion

We present Material Anything, a fully-automated, unified diffusion framework designed to generate physically-based materials for 3D objects. Unlike existing methods that rely on complex pipelines or case-specific optimizations, Material Anything offers a robust, end-to-end solution adaptable to objects under diverse lighting conditions. Our approach leverages a pre-trained image diffusion model, enhanced with a triple-head architecture and rendering loss to improve stability and material quality. Additionally, we introduce confidence masks as a dynamic switcher within the diffusion model, enabling it to effectively handle both textured and texture-less objects across varying lighting conditions. By employing a progressive material generation strategy guided by these confidence masks, along with a UV-space material refiner, our method ensures consistent, UV-ready material outputs. Extensive experiments demonstrate our approach outperforms existing methods across a wide range of object categories and lighting conditions.

cs.CV↗
arXiv2024

PPLqa: An Unsupervised Information-Theoretic Quality Metric for Comparing Generative Large Language Models

We propose PPLqa, an easy to compute, language independent, information-theoretic metric to measure the quality of responses of generative Large Language Models (LLMs) in an unsupervised way, without requiring ground truth annotations or human supervision. The method and metric enables users to rank generative language models for quality of responses, so as to make a selection of the best model for a given task. Our single metric assesses LLMs with an approach that subsumes, but is not explicitly based on, coherence and fluency (quality of writing) and relevance and consistency (appropriateness of response) to the query. PPLqa performs as well as other related metrics, and works better with long-form Q\&A. Thus, PPLqa enables bypassing the lengthy annotation process required for ground truth evaluations, and it also correlates well with human and LLM rankings.

cs.CL↗
arXiv2024

Improved Maximin Share Approximations for Chores by Bin Packing

We study fair division of indivisible chores among $n$ agents with additive cost functions using the popular fairness notion of maximin share (MMS). Since MMS allocations do not always exist for more than two agents, the goal has been to improve its approximations and identify interesting special cases where MMS allocations exists. We show the existence of 1) 1-out-of-$\lfloor \frac{9}{11}n\rfloor$ MMS allocations, which improves the state-of-the-art factor of 1-out-of-$\lfloor \frac{3}{4}n\rfloor$. 2) MMS allocations for factored instances, which resolves an open question posed by Ebadian et al. (2021). 3) $15/13$-MMS allocations for personalized bivalued instances, improving the state-of-the-art factor of $13/11$. We achieve these results by leveraging the HFFD algorithm of Huang and Lu (2021). Our approach also provides polynomial-time algorithms for computing an MMS allocation for factored instances and a $15/13$-MMS allocation for personalized bivalued instances.

cs.GT↗
arXiv2024

AutoCheck: Automatically Identifying Variables for Checkpointing by Data Dependency Analysis

Checkpoint/Restart (C/R) has been widely deployed in numerous HPC systems, Clouds, and industrial data centers, which are typically operated by system engineers. Nevertheless, there is no existing approach that helps system engineers without domain expertise, and domain scientists without system fault tolerance knowledge identify those critical variables accounted for correct application execution restoration in a failure for C/R. To address this problem, we propose an analytical model and a tool (AutoCheck) that can automatically identify critical variables to checkpoint for C/R. AutoCheck relies on first, analytically tracking and optimizing data dependency between variables and other application execution state, and second, a set of heuristics that identify critical variables for checkpointing from the refined data dependency graph (DDG). AutoCheck allows programmers to pinpoint critical variables to checkpoint quickly within a few minutes. We evaluate AutoCheck on 14 representative HPC benchmarks, demonstrating that AutoCheck can efficiently identify correct critical variables to checkpoint.

cs.SE↗
arXiv2024

Self-Supervised Graph Neural Networks for Enhanced Feature Extraction in Heterogeneous Information Networks

This paper explores the applications and challenges of graph neural networks (GNNs) in processing complex graph data brought about by the rapid development of the Internet. Given the heterogeneity and redundancy problems that graph data often have, traditional GNN methods may be overly dependent on the initial structure and attribute information of the graph, which limits their ability to accurately simulate more complex relationships and patterns in the graph. Therefore, this study proposes a graph neural network model under a self-supervised learning framework, which can flexibly combine different types of additional information of the attribute graph and its nodes, so as to better mine the deep features in the graph data. By introducing a self-supervisory mechanism, it is expected to improve the adaptability of existing models to the diversity and complexity of graph data and improve the overall performance of the model.

cs.LG↗
arXiv2024

Weighted Envy-free Allocation with Subsidy

We consider the problem of fair allocation of indivisible items with subsidies when agents have weighted entitlements. After highlighting several important differences from the unweighted case, we present several results concerning weighted envy-freeability including general characterizations, algorithms for achieving and testing weighted envy-freeability, lower and upper bounds of the amount of subsidies for envy-freeable allocations, and algorithms for achieving weighted envy-freeability along with other properties.

cs.GT↗
arXiv2024

Math-PUMA: Progressive Upward Multimodal Alignment to Enhance Mathematical Reasoning

Multimodal Large Language Models (MLLMs) excel in solving text-based mathematical problems, but they struggle with mathematical diagrams since they are primarily trained on natural scene images. For humans, visual aids generally enhance problem-solving, but MLLMs perform worse as information shifts from textual to visual modality. This decline is mainly due to their shortcomings in aligning images and text. To tackle aforementioned challenges, we propose Math-PUMA, a methodology focused on Progressive Upward Multimodal Alignment. This approach is designed to improve the mathematical reasoning skills of MLLMs through a three-stage training process, with the second stage being the critical alignment stage. We first enhance the language model's mathematical reasoning capabilities with extensive set of textual mathematical problems. We then construct a multimodal dataset with varying degrees of textual and visual information, creating data pairs by presenting each problem in at least two forms. By leveraging the Kullback-Leibler (KL) divergence of next-token prediction distributions to align visual and textual modalities, consistent problem-solving abilities are ensured. Finally, we utilize multimodal instruction tuning for MLLMs with high-quality multimodal data. Experimental results on multiple mathematical reasoning benchmarks demonstrate that the MLLMs trained with Math-PUMA surpass most open-source MLLMs. Our approach effectively narrows the performance gap for problems presented in different modalities. The code and data are available at: \url{https://github.com/wwzhuang01/Math-PUMA}.

cs.CL↗
arXiv2024

A Reduction from Chores Allocation to Job Scheduling

We consider allocating indivisible chores among agents with different cost functions, such that all agents receive a cost of at most a constant factor times their maximin share. The state-of-the-art was presented in In EC 2021 by Huang and Lu. They presented a non-polynomial-time algorithm, called HFFD, that attains an 11/9 approximation, and a polynomial-time algorithm that attains a 5/4 approximation. In this paper, we show that HFFD can be reduced to an algorithm called MultiFit, developed by Coffman, Garey and Johnson in 1978 for makespan minimization in job scheduling. Using this reduction, we prove that the approximation ratio of HFFD is in fact equal to that of MultiFit, which is known to be 13/11 in general, 20/17 for n at most 7, and 15/13 for n=3. Moreover, we develop an algorithm for (13/11+epsilon)-maximin-share allocation for any epsilon>0, with run-time polynomial in the problem size and 1/epsilon. For n=3, we can improve the algorithm to find a 15/13-maximin-share allocation with run-time polynomial in the problem size. Thus, we have practical algorithms that attain the best known approximation to maximin-share chore allocation.

cs.GT↗
arXiv2024

MoE-LPR: Multilingual Extension of Large Language Models through Mixture-of-Experts with Language Priors Routing

Large Language Models (LLMs) are often English-centric due to the disproportionate distribution of languages in their pre-training data. Enhancing non-English language capabilities through post-pretraining often results in catastrophic forgetting of the ability of original languages. Previous methods either achieve good expansion with severe forgetting or slight forgetting with poor expansion, indicating the challenge of balancing language expansion while preventing forgetting. In this paper, we propose a method called MoE-LPR (Mixture-of-Experts with Language Priors Routing) to alleviate this problem. MoE-LPR employs a two-stage training approach to enhance the multilingual capability. First, the model is post-pretrained into a Mixture-of-Experts (MoE) architecture by upcycling, where all the original parameters are frozen and new experts are added. In this stage, we focus improving the ability on expanded languages, without using any original language data. Then, the model reviews the knowledge of the original languages with replay data amounting to less than 1% of post-pretraining, where we incorporate language priors routing to better recover the abilities of the original languages. Evaluations on multiple benchmarks show that MoE-LPR outperforms other post-pretraining methods. Freezing original parameters preserves original language knowledge while adding new experts preserves the learning ability. Reviewing with LPR enables effective utilization of multilingual knowledge within the parameters. Additionally, the MoE architecture maintains the same inference overhead while increasing total model parameters. Extensive experiments demonstrate MoE-LPR's effectiveness in improving expanded languages and preserving original language proficiency with superior scalability. Code and scripts are freely available at https://github.com/zjwang21/MoE-LPR.git.

cs.CL↗
arXiv2024

Sample Enrichment via Temporary Operations on Subsequences for Sequential Recommendation

Sequential recommendation leverages interaction sequences to predict forthcoming user behaviors, crucial for crafting personalized recommendations. However, the true preferences of a user are inherently complex and high-dimensional, while the observed data is merely a simplified and low-dimensional projection of the rich preferences, which often leads to prevalent issues like data sparsity and inaccurate model training. To learn true preferences from the sparse data, most existing works endeavor to introduce some extra information or design some ingenious models. Although they have shown to be effective, extra information usually increases the cost of data collection, and complex models may result in difficulty in deployment. Innovatively, we avoid the use of extra information or alterations to the model; instead, we fill the transformation space between the observed data and the underlying preferences with randomness. Specifically, we propose a novel model-agnostic and highly generic framework for sequential recommendation called sample enrichment via temporary operations on subsequences (SETO), which temporarily and separately enriches the transformation space via sequence enhancement operations with rationality constraints in training. The transformation space not only exists in the process from input samples to preferences but also in preferences to target samples. We highlight our SETO's effectiveness and versatility over multiple representative and state-of-the-art sequential recommendation models (including six single-domain sequential models and two cross-domain sequential models) across multiple real-world datasets (including three single-domain datasets, three cross-domain datasets and a large-scale industry dataset).

cs.IR↗
arXiv2024

Top-k Representative Search for Comparative Tree Summarization

Data summarization aims at utilizing a small-scale summary to represent massive datasets as a whole, which is useful for visualization and information sipped generation. However, most existing studies of hierarchical summarization only work on \emph{one single tree} by selecting $k$ representative nodes, which neglects an important problem of comparative summarization on two trees. In this paper, given two trees with the same topology structure and different node weights, we aim at finding $k$ representative nodes, where $k_1$ nodes summarize the common relationship between them and $k_2$ nodes highlight significantly different sub-trees meanwhile satisfying $k_1+k_2=k$. To optimize summarization results, we introduce a scaling coefficient for balancing the summary view between two sub-trees in terms of similarity and difference. Additionally, we propose a novel definition based on the Hellinger distance to quantify the node distribution difference between the sub-trees. We present a greedy algorithm SVDT to find high-quality results with approximation guaranteed in an efficient way. Furthermore, we explore an extension of our comparative summarization to handle two trees with different structures. Extensive experiments demonstrate the effectiveness and efficiency of our SVDT algorithm against existing summarization competitors.

cs.DB↗
arXiv2024

SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoning

We present SeaEval, a benchmark for multilingual foundation models. In addition to characterizing how these models understand and reason with natural language, we also investigate how well they comprehend cultural practices, nuances, and values. Alongside standard accuracy metrics, we investigate the brittleness of foundation models in the dimensions of semantics and multilinguality. Our analyses span both open-sourced and closed models, leading to empirical results across classic NLP tasks, reasoning, and cultural comprehension. Key findings indicate (1) Most models exhibit varied behavior when given paraphrased instructions. (2) Many models still suffer from exposure bias (e.g., positional bias, majority label bias). (3) For questions rooted in factual, scientific, and commonsense knowledge, consistent responses are expected across multilingual queries that are semantically equivalent. Yet, most models surprisingly demonstrate inconsistent performance on these queries. (4) Multilingually-trained models have not attained "balanced multilingual" capabilities. Our endeavors underscore the need for more generalizable semantic representations and enhanced multilingual contextualization. SeaEval can serve as a launchpad for more thorough investigations and evaluations for multilingual and multicultural scenarios.

cs.CL↗
arXiv2024

MT-HCCAR: Multi-Task Deep Learning with Hierarchical Classification and Attention-based Regression for Cloud Property Retrieval

In the realm of Earth science, effective cloud property retrieval, encompassing cloud masking, cloud phase classification, and cloud optical thickness (COT) prediction, remains pivotal. Traditional methodologies necessitate distinct models for each sensor instrument due to their unique spectral characteristics. Recent strides in Earth Science research have embraced machine learning and deep learning techniques to extract features from satellite datasets' spectral observations. However, prevailing approaches lack novel architectures accounting for hierarchical relationships among retrieval tasks. Moreover, considering the spectral diversity among existing sensors, the development of models with robust generalization capabilities over different sensor datasets is imperative. Surprisingly, there is a dearth of methodologies addressing the selection of an optimal model for diverse datasets. In response, this paper introduces MT-HCCAR, an end-to-end deep learning model employing multi-task learning to simultaneously tackle cloud masking, cloud phase retrieval (classification tasks), and COT prediction (a regression task). The MT-HCCAR integrates a hierarchical classification network (HC) and a classification-assisted attention-based regression network (CAR), enhancing precision and robustness in cloud labeling and COT prediction. Additionally, a comprehensive model selection method rooted in K-fold cross-validation, one standard error rule, and two introduced performance scores is proposed to select the optimal model over three simulated satellite datasets OCI, VIIRS, and ABI. The experiments comparing MT-HCCAR with baseline methods, the ablation studies, and the model selection affirm the superiority and the generalization capabilities of MT-HCCAR.

cs.LG↗