arXiv subjects

Yueqi Wang

Publications and source records attributed to Yueqi Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Adaptive Crouzeix-Raviart finite elements for the first eigenpair of $p$-Laplacian

In this paper, we propose and analyze an adaptive Crouzeix-Raviart finite element method for computing the first Dirichlet eigenpair of the $p$-Laplacian problem. We prove that the sequence of error estimators produced by the adaptive algorithm has a vanishing limit and that, starting from a fine initial mesh, the relevant sequence of approximate eigenvalues converges to the first eigenvalue and the distance in a mesh-dependent broken norm between discrete eigenfunctions and the set composed of relevant continuous eigenfunctions also tends to zero. The analysis hinges on establishing a compactness property for Crouzeix-Raviart finite elements over a sequence of adaptively generated meshes, which represents key theoretical challenges and novelties. We present numerical results to illustrate the advantage of the proposed algorithm.

math.NA↗

arXiv2025

User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems

Exploration, the act of broadening user experiences beyond their established preferences, is challenging in large-scale recommendation systems due to feedback loops and limited signals on user exploration patterns. Large Language Models (LLMs) offer potential solutions by leveraging their world knowledge to recommend novel content outside these loops. A key challenge is aligning LLMs with user preferences while preserving their knowledge and reasoning. To enhance planning for new user interests using LLMs, this paper introduces a novel approach that combines hierarchical planning with LLM inference-time scaling. This method aims to improve recommendation relevancy without compromising novelty. We decouple novelty and user-alignment, training separate LLMs for each objective. We then scale up the novelty-focused LLM's inference and select the best-of-n predictions using the user-aligned LLM. Live experiments demonstrate efficacy, showing significant gains in both user satisfaction (measured by watch activity and active user counts) and exploration diversity.

cs.IR↗

arXiv2025

Your Causal Self-Attentive Recommender Hosts a Lonely Neighborhood

In the context of sequential recommendation, a pivotal issue pertains to the comparative analysis between bi-directional/auto-encoding (AE) and uni-directional/auto-regressive (AR) attention mechanisms, where the conclusions regarding architectural and performance superiority remain inconclusive. Previous efforts in such comparisons primarily involve summarizing existing works to identify a consensus or conducting ablation studies on peripheral modeling techniques, such as choices of loss functions. However, far fewer efforts have been made in (1) theoretical and (2) extensive empirical analysis of the self-attention module, the very pivotal structure on which performance and designing insights should be anchored. In this work, we first provide a comprehensive theoretical analysis of AE/AR attention matrix in the aspect of (1) sparse local inductive bias, a.k.a neighborhood effects, and (2) low rank approximation. Analytical metrics reveal that the AR attention exhibits sparse neighborhood effects suitable for generally sparse recommendation scenarios. Secondly, to support our theoretical analysis, we conduct extensive empirical experiments on comparing AE/AR attention on five popular benchmarks with AR performing better overall. Empirical results reported are based on our experimental pipeline named Modularized Design Space for Self-Attentive Recommender (ModSAR), supporting adaptive hyperparameter tuning, modularized design space and HuggingFace plug-ins. We invite the recommendation community to utilize/contribute to ModSAR to (1) conduct more module/model-level examining beyond AE/AR comparison and (2) accelerate state-of-the-art model design. Lastly, we shed light on future design choices for performant self-attentive recommenders. We make our pipeline implementation and data available at https://github.com/yueqirex/SAR-Check.

cs.IR↗

arXiv2024

Predicting band structures for 2D Photonic Crystals via Deep Learning

Photonic crystals (PhCs) are periodic dielectric structures that exhibit unique electromagnetic properties, such as the creation of band gaps where electromagnetic wave propagation is inhibited. Accurately predicting dispersion relations, which describe the frequency and direction of wave propagation, is vital for designing innovative photonic devices. However, traditional numerical methods, like the Finite Element Method (FEM), can encounter significant computational challenges due to the multiple scales present in photonic crystals, especially when calculating band structures across the entire Brillouin zone. To address this, we propose a supervised learning approach utilizing U-Net, along with transfer learning and Super-Resolution techniques, to forecast dispersion relations for 2D PhCs. Our model reduces computational expenses by producing high-resolution band structures from low-resolution data, eliminating the necessity for fine meshes throughout the Brillouin zone. The U-Net architecture enables the simultaneous prediction of multiple band functions, enhancing efficiency and accuracy compared to existing methods that handle each band function independently. Our findings demonstrate that the proposed model achieves high accuracy in predicting the initial band functions of 2D PhCs, while also significantly enhancing computational efficiency. This amalgamation of data-driven and traditional numerical techniques provides a robust framework for expediting the design and optimization of photonic crystals. The approach underscores the potential of integrating deep learning with established computational physics methods to tackle intricate multiscale problems, establishing a new benchmark for future PhC research and applications.

math.NA↗

arXiv2024

Train Once, Deploy Anywhere: Matryoshka Representation Learning for Multimodal Recommendation

Despite recent advancements in language and vision modeling, integrating rich multimodal knowledge into recommender systems continues to pose significant challenges. This is primarily due to the need for efficient recommendation, which requires adaptive and interactive responses. In this study, we focus on sequential recommendation and introduce a lightweight framework called full-scale Matryoshka representation learning for multimodal recommendation (fMRLRec). Our fMRLRec captures item features at different granularities, learning informative representations for efficient recommendation across multiple dimensions. To integrate item features from diverse modalities, fMRLRec employs a simple mapping to project multimodal item features into an aligned feature space. Additionally, we design an efficient linear transformation that embeds smaller features into larger ones, substantially reducing memory requirements for large-scale training on recommendation data. Combined with improved state space modeling techniques, fMRLRec scales to different dimensions and only requires one-time training to produce multiple models tailored to various granularities. We demonstrate the effectiveness and efficiency of fMRLRec on multiple benchmark datasets, which consistently achieves superior performance over state-of-the-art baseline methods. We make our code and data publicly available at https://github.com/yueqirex/fMRLRec.

cs.IR↗

arXiv2024

Fast switchable unidirectional magnon emitter

Magnon spintronics is an emerging field that explores the use of magnons, the quanta of spin waves in magnetic materials for information processing and communication. Achieving unidirectional information transport with fast switching capability is critical for the development of fast integrated magnonic circuits, which offer significant advantages in high-speed, low-power information processing. However, previous unidirectional information transport has primarily focused on Damon-Eshbach spin wave modes, which are non-switchable as their propagation direction is defined by the direction of the external field and cannot be changed in a short time. Here, we experimentally demonstrate a fast switchable unidirectional magnon emitter in the forward volume spin wave mode by a current-induced asymmetric Oersted field. Our findings reveal significant nonreciprocity and nanosecond switchability, underscoring the potential of the method to advance high-speed spin-wave processing networks.

cond-mat.mes-hall↗

arXiv2024

On Edge Multiscale Space based Hybrid Schwarz Preconditioner for Helmholtz Problems with Large Wavenumbers

In this work, we develop a novel hybrid Schwarz method, termed as edge multiscale space based hybrid Schwarz (EMs-HS), for solving the Helmholtz problem with large wavenumbers. The problem is discretized using $H^1$-conforming nodal finite element methods on meshes of size $h$ decreasing faster than $k^{-1}$ such that the discretization error remains bounded as the wavenumber $k$ increases. EMs-HS consists of a one-level Schwarz preconditioner (RAS-imp) and a coarse solver in a multiplicative way. The RAS-imp preconditioner solves local problems on overlapping subdomains with impedance boundary conditions in parallel, and combines the local solutions using partition of unity. The coarse space is an edge multiscale space proposed in [13]. The key idea is to first establish a local splitting of the solution over each subdomain by a local bubble part and local Helmholtz harmonic extension part, and then to derive a global splitting by means of the partition of unity. This facilitates representing the solution as the sum of a global bubble part and a global Helmholtz harmonic extension part. We prove that the EMs-HS preconditioner leads to a convergent fixed-point iteration uniformly for large wavenumbers, by rigorously analyzing the approximation properties of the coarse space to the global Helmholtz harmonic extension part and to the solution of the adjoint problem. Distinctly, the theoretical convergence analysis are valid in two extreme cases: using minimal overlapping size among subdomains (of order $h$), or using coarse spaces of optimal dimension (of magnitude $k^d$, where $d$ is the spatial dimension). We provide extensive numerical results on the sharpness of the theoretical findings and also demonstrate the method on challenging heterogeneous models.

math.NA↗

arXiv2024

Amortized Probabilistic Detection of Communities in Graphs

Learning community structures in graphs has broad applications across scientific domains. While graph neural networks (GNNs) have been successful in encoding graph structures, existing GNN-based methods for community detection are limited by requiring knowledge of the number of communities in advance, in addition to lacking a proper probabilistic formulation to handle uncertainty. We propose a simple framework for amortized community detection, which addresses both of these issues by combining the expressive power of GNNs with recent methods for amortized clustering. Our models consist of a graph representation backbone that extracts structural information and an amortized clustering network that naturally handles variable numbers of clusters. Both components combine into well-defined models of the posterior distribution of graph communities and are jointly optimized given labeled graphs. At inference time, the models yield parallel samples from the posterior of community labels, quantifying uncertainty in a principled way. We evaluate several models from our framework on synthetic and real datasets, and demonstrate improved performance compared to previous methods. As a separate contribution, we extend recent amortized probabilistic clustering architectures by adding attention modules, which yield further improvements on community detection tasks.

stat.ML↗

arXiv2024

An hp-Adaptive Sampling Algorithm for Dispersion Relation Reconstruction of 3D Photonic Crystals

In this work we investigate the computation of dispersion relation (i.e., band functions) for three-dimensional photonic crystals, formulated as a parameterized Maxwell eigenvalue problem, using a novel hp-adaptive sampling algorithm. We develop an adaptive sampling algorithm in the parameter domain such that local elements with singular points are refined at each iteration, construct a conforming element-wise polynomial space on the adaptive mesh such that the distribution of the local polynomial spaces reflects the regularity of the band functions, and define an element-wise Lagrange interpolation operator to approximate the band functions. We rigorously prove the convergence of the algorithm. To illustrate the significant potential of the algorithm, we present two numerical tests with band gap optimization.

math.NA↗

arXiv2024

LLMs for User Interest Exploration in Large-scale Recommendation Systems

Traditional recommendation systems are subject to a strong feedback loop by learning from and reinforcing past user-item interactions, which in turn limits the discovery of novel user interests. To address this, we introduce a hybrid hierarchical framework combining Large Language Models (LLMs) and classic recommendation models for user interest exploration. The framework controls the interfacing between the LLMs and the classic recommendation models through "interest clusters", the granularity of which can be explicitly determined by algorithm designers. It recommends the next novel interests by first representing "interest clusters" using language, and employs a fine-tuned LLM to generate novel interest descriptions that are strictly within these predefined clusters. At the low level, it grounds these generated interests to an item-level policy by restricting classic recommendation models, in this case a transformer-based sequence recommender to return items that fall within the novel clusters generated at the high level. We showcase the efficacy of this approach on an industrial-scale commercial platform serving billions of users. Live experiments show a significant increase in both exploration of novel interests and overall user enjoyment of the platform.

cs.IR↗

arXiv2024

Improving Multi-Instance GPU Efficiency via Sub-Entry Sharing TLB Design

NVIDIA's Multi-Instance GPU (MIG) technology enables partitioning GPU computing power and memory into separate hardware instances, providing complete isolation including compute resources, caches, and memory. However, prior work identifies that MIG does not extend to partitioning the last-level TLB (i.e., L3 TLB), which remains shared among all instances. To enhance TLB reach, NVIDIA GPUs reorganized the TLB structure with 16 sub-entries in each L3 TLB entry that have a one-to-one mapping to the address translations for 16 pages of size 64KB located within the same 1MB aligned range. Our comprehensive investigation of address translation efficiency in MIG identifies two main issues caused by L3 TLB sharing interference: (i) it results in performance degradation for co-running applications, and (ii) TLB sub-entries are not fully utilized before eviction. Based on this observation, we propose STAR to improve the utilization of TLB sub-entries through dynamic sharing of TLB entries across multiple base addresses. STAR evaluates TLB entries based on their sub-entry utilization to optimize address translation storage, dynamically adjusting between a shared and non-shared status to cater to current demand. We show that STAR improves overall performance by an average of 30.2% across various multi-tenant workloads.

cs.DC↗

arXiv2023

An hp-Adaptive Sampling Algorithm on Dispersion Relation Reconstruction for 2D Photonic Crystals

Computing the dispersion relation for two-dimensional photonic crystals is a notoriously challenging task: It involves solving parameterized Helmholtz eigenvalue problems with high-contrast coefficients. To resolve the challenge, we propose a novel hp-adaptive sampling scheme that can detect singular points via adaptive mesh refinement in the parameter domain, and meanwhile, allow for adaptively enriching the local polynomial spaces on the elements that do not contain singular points. In this way, we obtain an element-wise interpolation on an adaptive mesh. We derive an exponential convergence rate when the number of singular points is finite, and a first-order convergence rate otherwise. Numerical tests are provided to illustrate its performance.

math.NA↗

arXiv2023

Linear Recurrent Units for Sequential Recommendation

State-of-the-art sequential recommendation relies heavily on self-attention-based recommender models. Yet such models are computationally expensive and often too slow for real-time recommendation. Furthermore, the self-attention operation is performed at a sequence-level, thereby making low-cost incremental inference challenging. Inspired by recent advances in efficient language modeling, we propose linear recurrent units for sequential recommendation (LRURec). Similar to recurrent neural networks, LRURec offers rapid inference and can achieve incremental inference on sequential inputs. By decomposing the linear recurrence operation and designing recursive parallelization in our framework, LRURec provides the additional benefits of reduced model size and parallelizable training. Moreover, we optimize the architecture of LRURec by implementing a series of modifications to address the lack of non-linearity and improve training dynamics. To validate the effectiveness of our proposed LRURec, we conduct extensive experiments on multiple real-world datasets and compare its performance against state-of-the-art sequential recommenders. Experimental results demonstrate the effectiveness of LRURec, which consistently outperforms baselines by a significant margin. Results also highlight the efficiency of LRURec with our parallelized training paradigm and fast inference on long sequences, showing its potential to further enhance user experience in sequential recommendation.

cs.IR↗

arXiv2023

Dispersion relation reconstruction for 2D Photonic Crystals based on polynomial interpolation

Dispersion relation reflects the dependence of wave frequency on its wave vector when the wave passes through certain material. It demonstrates the properties of this material and thus it is critical. However, dispersion relation reconstruction is very time consuming and expensive. To address this bottleneck, we propose in this paper an efficient dispersion relation reconstruction scheme based on global polynomial interpolation for the approximation of 2D photonic band functions. Our method relies on the fact that the band functions are piecewise analytic with respect to the wave vector in the first Brillouin zone. We utilize suitable sampling points in the first Brillouin zone at which we solve the eigenvalue problem involved in the band function calculation, and then employ Lagrange interpolation to approximate the band functions on the whole first Brillouin zone. Numerical results show that our proposed methods can significantly improve the computational efficiency.

math.NA↗

arXiv2023

Quantized Low-Rank Multivariate Regression with Random Dithering

Low-rank multivariate regression (LRMR) is an important statistical learning model that combines highly correlated tasks as a multiresponse regression problem with low-rank priori on the coefficient matrix. In this paper, we study quantized LRMR, a practical setting where the responses and/or the covariates are discretized to finite precision. We focus on the estimation of the underlying coefficient matrix. To make consistent estimator that could achieve arbitrarily small error possible, we employ uniform quantization with random dithering, i.e., we add appropriate random noise to the data before quantization. Specifically, uniform dither and triangular dither are used for responses and covariates, respectively. Based on the quantized data, we propose the constrained Lasso and regularized Lasso estimators, and derive the non-asymptotic error bounds. With the aid of dithering, the estimators achieve minimax optimal rate, while quantization only slightly worsens the multiplicative factor in the error rate. Moreover, we extend our results to a low-rank regression model with matrix responses. We corroborate and demonstrate our theoretical results via simulations on synthetic data or image restoration.

stat.ML↗

arXiv2023

Learning from Negative User Feedback and Measuring Responsiveness for Sequential Recommenders

Sequential recommenders have been widely used in industry due to their strength in modeling user preferences. While these models excel at learning a user's positive interests, less attention has been paid to learning from negative user feedback. Negative user feedback is an important lever of user control, and comes with an expectation that recommenders should respond quickly and reduce similar recommendations to the user. However, negative feedback signals are often ignored in the training objective of sequential retrieval models, which primarily aim at predicting positive user interactions. In this work, we incorporate explicit and implicit negative user feedback into the training objective of sequential recommenders in the retrieval stage using a "not-to-recommend" loss function that optimizes for the log-likelihood of not recommending items with negative feedback. We demonstrate the effectiveness of this approach using live experiments on a large-scale industrial recommender system. Furthermore, we address a challenge in measuring recommender responsiveness to negative feedback by developing a counterfactual simulation framework to compare recommender responses between different user actions, showing improved responsiveness from the modeling change.

cs.IR↗

arXiv2023

Quasi liquid layer-pressure asymmetrical model for the motion of of a curling rock on ice surface

In this paper, we present a new model based on Quasi liquid layer to explain why the direction of lateral motion of the curling rock on ice surface is opposite to the other material surface. As we know, under the action of inertial force, the pressure on the ice surface in front of curling is greater than that on the back. So we assert that the firction coefficientin front of curling is lower than that on the bank under different pressure. In order to explain the pressure impact on friction coefficient, we qualitatively account for the reason why the coefficient of friction increases under the pressure and approximately calculated the relationship between the pressure and the thickness of the quasi_liquid layer on the ice surface. Then we calculate the function expression between temperature , pressure and the firction coefficient by the function between temperature and friction coefficient.

physics.geo-ph↗

arXiv2022

Searching for the charged-current non-standard neutrino interactions at the $e^{-}p$ colliders

Considering the theoretical constraints on the charged-current (CC) non-standard neutrino interaction (NSI) parameters in a simplified $W'$ model, we study the sensitivities of the Large Hadron electron Collider (LHeC) and Future Circular Collider-hadron electron (FCC-he) to the CC NSI parameters through the subprocess $e^{-}q \to W' \to v_{e}q'$. Our results show that the LHeC with $\mathcal{L} = 100\;{\rm fb}^{-1}$ is a little more sensitive to the CC NSI parameters than that of the high-luminosity (HL) LHC, and the absolute values of the lower limits on the CC NSI parameters at the FCC-he are smaller than those of the HL-LHC about two orders of magnitude.

hep-ph↗