arXiv subjects

Wei Hu

Publications and source records attributed to Wei Hu.

At least 91 records · Page 5Linked to original sources

arXiv2025

Non-Hermitian Bethe-Salpeter Equation for Open Systems: Emergence of Exceptional Points in Excitonic Spectra from First Principles

In open quantum systems hosting excitons, dissipation mechanisms critically shape the excitonic dynamics, band-structure and topological properties. A microscopic understanding of excitons in such non-Hermitian settings demands a first-principles generalization of the Bethe-Salpeter equation (BSE). Building on a recently introduced nonequilibrium Green's function formalism compatible with Lindbladian dynamics, we derive a non-Hermitian BSE from diagrammatic perturbation theory on the Keldysh contour, and obtain a microscopic excitonic Hamiltonian that incorporates dissipation while preserving causality. We apply the formalism to valley excitons in transition metal dichalcogenides coupled to structured photon baths. We uncover a rich landscape of exceptional points in momentum space, forming either discrete sets or continuous manifolds, depending on bath structure. The exceptional points give rise to nonanalytic valley-polarization, unusual polarization pattern in photoluminescence, and nontrivial topological signatures. Our results establish a first-principles framework for predicting and controlling excitonic behavior in open quantum materials, showing how engineered environments can be leveraged to induce and manipulate non-Hermitian and topological properties.

cond-mat.mes-hall↗

arXiv2025

A Universal Deep Learning Force Field for Molecular Dynamic Simulation and Vibrational Spectra Prediction

Accurate and efficient simulation of infrared (IR) and Raman spectra is essential for molecular identification and structural analysis. Traditional quantum chemistry methods based on the harmonic approximation neglect anharmonicity and nuclear quantum effects, while ab initio molecular dynamics (AIMD) remains computationally expensive. Here, we integrate our deep equivariant tensor attention network (DetaNet) with a velocity-Verlet integrator to enable fast and accurate machine learning molecular dynamics (MLMD) simulations for spectral prediction. Trained on the QMe14S dataset containing energies, forces, dipole moments, and polarizabilities for 186,102 small organic molecules, DetaNet yields a universal and transferable force field with high-order tensor prediction capability. Using time-correlation functions derived from MLMD and ring-polymer molecular dynamics (RPMD) trajectories, we computed IR and Raman spectra that accurately reproduce anharmonic and nuclear quantum effects. Benchmark tests on isolated molecules, including polycyclic aromatic hydrocarbons, demonstrate that the DetaNet-based MD approach achieves near-experimental spectral accuracy with speedups up to three orders of magnitude over AIMD. Furthermore, the framework extends seamlessly to molecular and inorganic crystals, molecular aggregates, and biological macromolecules such as polypeptides with minimal fine-tuning. In all systems, DetaNet maintains high accuracy while significantly reducing computational cost. Overall, this work establishes a universal machine learning force field and tensor-aware MLMD framework that enable fast, accurate, and broadly applicable dynamic simulations and IR/Raman spectral predictions across diverse molecular and material systems.

physics.chem-ph↗

arXiv2025

Learning Compact Representations of LLM Abilities via Item Response Theory

Recent years have witnessed a surge in the number of large language models (LLMs), yet efficiently managing and utilizing these vast resources remains a significant challenge. In this work, we explore how to learn compact representations of LLM abilities that can facilitate downstream tasks, such as model routing and performance prediction on new benchmarks. We frame this problem as estimating the probability that a given model will correctly answer a specific query. Inspired by the item response theory (IRT) in psychometrics, we model this probability as a function of three key factors: (i) the model's multi-skill ability vector, (2) the query's discrimination vector that separates models of differing skills, and (3) the query's difficulty scalar. To learn these parameters jointly, we introduce a Mixture-of-Experts (MoE) network that couples model- and query-level embeddings. Extensive experiments demonstrate that our approach leads to state-of-the-art performance in both model routing and benchmark accuracy prediction. Moreover, analysis validates that the learned parameters encode meaningful, interpretable information about model capabilities and query characteristics.

cs.AI↗

arXiv2025

WoW: A Window-to-Window Incremental Index for Range-Filtering Approximate Nearest Neighbor Search

Given a hybrid dataset where every data object consists of a vector and an attribute value, for each query with a target vector and a range filter, range-filtering approximate nearest neighbor search (RFANNS) aims to retrieve the most similar vectors from the dataset and the corresponding attribute values fall in the query range. It is a fundamental function in vector database management systems and intelligent systems with embedding abilities. Dedicated indices for RFANNS accelerate query speed with an acceptable accuracy loss on nearest neighbors. However, they are still facing the challenges to be constructed incrementally and generalized to achieve superior query performance for arbitrary range filters. In this paper, we introduce a window graph-based RFANNS index. For incremental construction, we propose an insertion algorithm to add new vector-attribute pairs into hierarchical window graphs with varying window size. To handle arbitrary range filters, we optimize relevant window search for attribute filter checks and vector distance computations by range selectivity. Extensive experiments on real-world datasets show that for index construction, the indexing time is on par with the most building-efficient index, and 4.9x faster than the most query-efficient index with 0.4-0.5x smaller size; For RFANNS query, it is 4x faster than the most efficient incremental index, and matches the performance of the best statically-built index.

cs.DB↗

arXiv2025

RISE: Enhancing VLM Image Annotation with Self-Supervised Reasoning

Vision-Language Models (VLMs) struggle with complex image annotation tasks, such as emotion classification and context-driven object detection, which demand sophisticated reasoning. Standard Supervised Fine-Tuning (SFT) focuses solely on annotation outcomes, ignoring underlying rationales, while Visual Reinforcement Fine-Tuning (Visual-RFT) produces inconsistent Chains of Thought (CoTs) due to the absence of high-quality, verified CoTs during pre-training. We introduce RISE (Reason-Inspire-Strengthen-Expertise), a two-stage framework to overcome these limitations. In the Reason stage (RISE-CoT), a reinforcement learning-driven "annotation-reasoning-annotation" closed-loop generates visually grounded, logically consistent CoTs by verifying their ability to reconstruct original annotations without direct leakage. The Inspire and Strengthen stage (RISE-R1) leverages a high-quality CoT subset, filtered by RISE-CoT rewards, for supervised fine-tuning, followed by reinforcement fine-tuning to produce interpretable reasoning and accurate annotations, achieving Expertise in complex visual tasks. Evaluated on complex and simple image annotation tasks, RISE-trained Qwen2-VL-2B outperforms SFT and Visual-RFT, achieving robust performance and enhanced explainability. RISE offers a self-supervised solution for advancing VLM reasoning without requiring manually annotated CoTs.Code and resources are available at: https://github.com/HSH55/RISE.

cs.LG↗

arXiv2025

Avoiding Knowledge Edit Skipping in Multi-hop Question Answering with Guided Decomposition

In a rapidly evolving world where information updates swiftly, knowledge in large language models (LLMs) becomes outdated quickly. Retraining LLMs is not a cost-effective option, making knowledge editing (KE) without modifying parameters particularly necessary. We find that although existing retrieval-augmented generation (RAG)-based KE methods excel at editing simple knowledge, they struggle with KE in multi-hop question answering due to the issue of "edit skipping", which refers to skipping the relevant edited fact in inference. In addition to the diversity of natural language expressions of knowledge, edit skipping also arises from the mismatch between the granularity of LLMs in problem-solving and the facts in the edited memory. To address this issue, we propose a novel Iterative Retrieval-Augmented Knowledge Editing method with guided decomposition (IRAKE) through the guidance from single edited facts and entire edited cases. Experimental results demonstrate that IRAKE mitigates the failure of editing caused by edit skipping and outperforms state-of-the-art methods for KE in multi-hop question answering.

cs.CL↗

arXiv2025

Sparse Mean Estimation in Adversarial Settings via Incremental Learning

In this paper, we study the problem of sparse mean estimation under adversarial corruptions, where the goal is to estimate the $k$-sparse mean of a heavy-tailed distribution from samples contaminated by adversarial noise. Existing methods face two key limitations: they require prior knowledge of the sparsity level $k$ and scale poorly to high-dimensional settings. We propose a simple and scalable estimator that addresses both challenges. Specifically, it learns the $k$-sparse mean without knowing $k$ in advance and operates in near-linear time and memory with respect to the ambient dimension. Under a moderate signal-to-noise ratio, our method achieves the optimal statistical rate, matching the information-theoretic lower bound. Extensive simulations corroborate our theoretical guarantees. At the heart of our approach is an incremental learning phenomenon: we show that a basic subgradient method applied to a nonconvex two-layer formulation with an $\ell_1$-loss can incrementally learn the $k$ nonzero components of the true mean while suppressing the rest. More broadly, our work is the first to reveal the incremental learning phenomenon of the subgradient method in the presence of heavy-tailed distributions and adversarial corruption.

cs.LG↗

arXiv2025

Complexity Analysis of the Regular Simplicial Search Method with Reflection and Shrinking Steps for Derivative-Free Optimization

Simplex-type methods, such as the well-known Nelder-Mead algorithm, are widely used in derivative-free optimization (DFO), particularly in practice. Despite their popularity, the theoretical understanding of their convergence properties has been limited, and until very recently essentially no worst-case complexity bounds were available. Recently, Cao et al. provided a sharp error bound for linear interpolation and extrapolation and derived a worst-case complexity result for a basic simplex-type method. Motivated by this, we propose a practical and provable algorithm -- the regular simplicial search method (RSSM), that incorporates reflection and shrinking steps, akin to the original method of Spendley et al. We establish worst-case complexity bounds in nonconvex, convex, and strongly convex cases. These results provide guarantees on convergence rates and lay the groundwork for future complexity analysis of more advanced simplex-type algorithms.

math.OC↗

arXiv2025

HierOctFusion: Multi-scale Octree-based 3D Shape Generation via Part-Whole-Hierarchy Message Passing

3D content generation remains a fundamental yet challenging task due to the inherent structural complexity of 3D data. While recent octree-based diffusion models offer a promising balance between efficiency and quality through hierarchical generation, they often overlook two key insights: 1) existing methods typically model 3D objects as holistic entities, ignoring their semantic part hierarchies and limiting generalization; and 2) holistic high-resolution modeling is computationally expensive, whereas real-world objects are inherently sparse and hierarchical, making them well-suited for layered generation. Motivated by these observations, we propose HierOctFusion, a part-aware multi-scale octree diffusion model that enhances hierarchical feature interaction for generating fine-grained and sparse object structures. Furthermore, we introduce a cross-attention conditioning mechanism that injects part-level information into the generation process, enabling semantic features to propagate effectively across hierarchical levels from parts to the whole. Additionally, we construct a 3D dataset with part category annotations using a pre-trained segmentation model to facilitate training and evaluation. Experiments demonstrate that HierOctFusion achieves superior shape quality and efficiency compared to prior methods.

cs.CV↗

arXiv2025

A GPU-Accelerated Three-Dimensional Crack Element Method for Transient Dynamic Fracture Simulation

This work presents a novel three-dimensional Crack Element Method (CEM) designed to model transient dynamic crack propagation in quasi-brittle materials efficiently. CEM introduces an advanced element-splitting algorithm that enables element-wise crack growth, including crack branching. Based on the evolving topology of split elements, an original formulation for computing the fracture energy release rate in three dimensions is derived. A series of benchmark examples is conducted to demonstrate that the proposed 3D CEM accurately simulates both single crack propagation and complex crack branching scenarios. Furthermore, all three-dimensional simulations are GPU-accelerated, achieving high levels of computational efficiency, consistency, and accuracy.

cs.CE↗

arXiv2025

Hypercube-Based Retrieval-Augmented Generation for Scientific Question-Answering

Large language models (LLMs) often need to incorporate external knowledge to solve theme-specific problems. Retrieval-augmented generation (RAG) has shown its high promise, empowering LLMs to generate more qualified responses with retrieved external data and knowledge. However, most RAG methods retrieve relevant documents based on either sparse or dense retrieval methods or their combinations, which overlooks the essential, multi-dimensional, and structured semantic information present in documents. This structured information plays a critical role in finding concise yet highly relevant information for domain knowledge-intensive tasks, such as scientific question-answering (QA). In this work, we introduce a multi-dimensional (cube) structure, Hypercube, which can index and allocate documents in a pre-defined multi-dimensional space. Built on the hypercube, we further propose Hypercube-RAG, a novel RAG framework for precise and efficient retrieval. Given a query, Hypercube-RAG first decomposes it based on its entities, phrases, and topics along with pre-defined hypercube dimensions, and then retrieves relevant documents from cubes by aligning these decomposed components with corresponding dimensions. Experiments on three datasets across different domains demonstrate that our method improves response accuracy by 3.7% and retrieval accuracy by 5.3% over the strongest RAG baseline. It also boosts retrieval efficiency (speed) by one or two magnitudes faster than graph-based RAG. Notably, our Hypercube-RAG inherently offers explainability by revealing those underlying dimensions used for retrieval. The code and data are available at https://github.com/JimengShi/Hypercube-RAG.

cs.LG↗

arXiv2025

Provable Low-Frequency Bias of In-Context Learning of Representations

In-context learning (ICL) enables large language models (LLMs) to acquire new behaviors from the input sequence alone without any parameter updates. Recent studies have shown that ICL can surpass the original meaning learned in pretraining stage through internalizing the structure the data-generating process (DGP) of the prompt into the hidden representations. However, the mechanisms by which LLMs achieve this ability is left open. In this paper, we present the first rigorous explanation of such phenomena by introducing a unified framework of double convergence, where hidden representations converge both over context and across layers. This double convergence process leads to an implicit bias towards smooth (low-frequency) representations, which we prove analytically and verify empirically. Our theory explains several open empirical observations, including why learned representations exhibit globally structured but locally distorted geometry, and why their total energy decays without vanishing. Moreover, our theory predicts that ICL has an intrinsic robustness towards high-frequency noise, which we empirically confirm. These results provide new insights into the underlying mechanisms of ICL, and a theoretical foundation to study it that hopefully extends to more general data distributions and settings.

cs.LG↗

arXiv2025

Conflict Detection for Temporal Knowledge Graphs:A Fast Constraint Mining Algorithm and New Benchmarks

Temporal facts, which are used to describe events that occur during specific time periods, have become a topic of increased interest in the field of knowledge graph (KG) research. In terms of quality management, the introduction of time restrictions brings new challenges to maintaining the temporal consistency of KGs. Previous studies rely on manually enumerated temporal constraints to detect conflicts, which are labor-intensive and may have granularity issues. To address this problem, we start from the common pattern of temporal facts and propose a pattern-based temporal constraint mining method, PaTeCon. Unlike previous studies, PaTeCon uses graph patterns and statistical information relevant to the given KG to automatically generate temporal constraints, without the need for human experts. In this paper, we illustrate how this method can be optimized to achieve significant speed improvement. We also annotate Wikidata and Freebase to build two new benchmarks for conflict detection. Extensive experiments demonstrate that our pattern-based automatic constraint mining approach is highly effective in generating valuable temporal constraints.

cs.AI↗

arXiv2025

Learning Free Terminal Time Optimal Closed-loop Control of Manipulators

This paper presents a novel approach to learning free terminal time closed-loop control for robotic manipulation tasks, enabling dynamic adjustment of task duration and control inputs to enhance performance. We extend the supervised learning approach, namely solving selected optimal open-loop problems and utilizing them as training data for a policy network, to the free terminal time scenario. Three main challenges are addressed in this extension. First, we introduce a marching scheme that enhances the solution quality and increases the success rate of the open-loop solver by gradually refining time discretization. Second, we extend the QRnet in Nakamura-Zimmerer et al. (2021b) to the free terminal time setting to address discontinuity and improve stability at the terminal state. Third, we present a more automated version of the initial value problem (IVP) enhanced sampling method from previous work (Zhang et al., 2022) to adaptively update the training dataset, significantly improving its quality. By integrating these techniques, we develop a closed-loop policy that operates effectively over a broad domain with varying optimal time durations, achieving near globally optimal total costs.

math.OC↗

arXiv2025

Fast3D: Accelerating 3D Multi-modal Large Language Models for Efficient 3D Scene Understanding

While 3D Multi-modal Large Language Models (MLLMs) demonstrate remarkable scene understanding capabilities, their practical deployment faces critical challenges due to computational inefficiency. The key bottleneck stems from processing excessive object-centric visual tokens required for comprehensive 3D scene representation. Although visual token pruning has shown promise in accelerating 2D MLLMs, its applicability to 3D domains remains largely unexplored due to fundamental disparities in token structures. In this paper, we reveal two critical insights: (1) Significant redundancy exists in object-level 3D token representations, analogous to patch-level redundancy in 2D systems; (2) Global attention patterns exhibit strong predictive power for identifying non-essential tokens in 3D contexts. Building on these observations, we propose Fast3D, a plug-and-play visual token pruning framework for 3D MLLMs featuring two technical innovations: (1) Global Attention Prediction (GAP), where a lightweight neural network learns to predict the global attention distributions of the target model, enabling efficient token importance estimation for precise pruning guidance; (2) Sample-Adaptive visual token Pruning (SAP), which introduces dynamic token budgets through attention-based complexity assessment, automatically adjusting layer-wise pruning ratios based on input characteristics. Both of these two techniques operate without modifying the parameters of the target model. Extensive evaluations across five benchmarks validate the effectiveness of Fast3D, particularly under high visual token pruning ratios. Code is available at https://github.com/wencan25/Fast3D

cs.CV↗

arXiv2025

Behave Your Motion: Habit-preserved Cross-category Animal Motion Transfer

Animal motion embodies species-specific behavioral habits, making the transfer of motion across categories a critical yet complex task for applications in animation and virtual reality. Existing motion transfer methods, primarily focused on human motion, emphasize skeletal alignment (motion retargeting) or stylistic consistency (motion style transfer), often neglecting the preservation of distinct habitual behaviors in animals. To bridge this gap, we propose a novel habit-preserved motion transfer framework for cross-category animal motion. Built upon a generative framework, our model introduces a habit-preservation module with category-specific habit encoder, allowing it to learn motion priors that capture distinctive habitual characteristics. Furthermore, we integrate a large language model (LLM) to facilitate the motion transfer to previously unobserved species. To evaluate the effectiveness of our approach, we introduce the DeformingThings4D-skl dataset, a quadruped dataset with skeletal bindings, and conduct extensive experiments and quantitative analyses, which validate the superiority of our proposed model.

cs.CV↗

arXiv2025

HyperGCT: A Dynamic Hyper-GNN-Learned Geometric Constraint for 3D Registration

Geometric constraints between feature matches are critical in 3D point cloud registration problems. Existing approaches typically model unordered matches as a consistency graph and sample consistent matches to generate hypotheses. However, explicit graph construction introduces noise, posing great challenges for handcrafted geometric constraints to render consistency. To overcome this, we propose HyperGCT, a flexible dynamic Hyper-GNN-learned geometric ConstrainT that leverages high-order consistency among 3D correspondences. To our knowledge, HyperGCT is the first method that mines robust geometric constraints from dynamic hypergraphs for 3D registration. By dynamically optimizing the hypergraph through vertex and edge feature aggregation, HyperGCT effectively captures the correlations among correspondences, leading to accurate hypothesis generation. Extensive experiments on 3DMatch, 3DLoMatch, KITTI-LC, and ETH show that HyperGCT achieves state-of-the-art performance. Furthermore, HyperGCT is robust to graph noise, demonstrating a significant advantage in terms of generalization.

cs.CV↗

arXiv2025

Deep Picard Iteration for High-Dimensional Nonlinear PDEs

We present the Deep Picard Iteration (DPI) method, a new deep learning approach for solving high-dimensional partial differential equations (PDEs). The core innovation of DPI lies in its use of Picard iteration to reformulate the typically complex training objectives of neural network-based PDE solutions into much simpler, standard regression tasks based on function values and gradients. This design not only greatly simplifies the optimization process but also offers the potential for further scalability through parallel data generation. Crucially, to fully realize the benefits of regressing on both function values and gradients in the DPI method, we address the issue of infinite variancein the estimators of gradients by incorporating a control variate, supported by our theoretical analysis. Our experiments on problems up to 100 dimensions demonstrate that DPI consistently outperforms existing state-of-the-art methods, with greater robustness to hyperparameters, particularly in challenging scenarios with long time horizons and strong nonlinearity. The code is available at https://github.com/DeepOptimalControl/DeepPicardIteration.

math.NA↗