检索结果-内蒙古大学图书馆

A Query Optimization Method Utilizing Large Language Models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Yao, Zhiming Li, Haoyang Zhang, Jing Li, Cuiping Chen, Hong School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering MOE China Engineering Research Center of Database and Business Intelligence MOE China

Query optimization is a critical task in database systems, focused on determining the most efficient way to execute a query from an enormous set of possible strategies. Traditional approaches rely on heuristic search methods and cost predictions, but these often struggle with the complexity of the search space and inaccuracies in performance estimation, leading to suboptimal plan choices. This paper presents LLMOpt, a novel framework that leverages Large Language Models (LLMs) to address these challenges through two innovative components: (1) LLM for Plan Candidate Generation (LLMOpt(G)), which eliminates heuristic search by utilizing the reasoning abilities of LLMs to directly generate high-quality query plans, and (2) LLM for Plan Candidate Selection (LLMOpt(S)), a list-wise cost model that compares candidates globally to enhance selection accuracy. To adapt LLMs for query optimization, we propose fine-tuning pre-trained models using optimization data collected offline. Experimental results on the JOB, JOB-EXT, and Stack benchmarks show that LLMOpt(G) and LLMOpt(S) outperform state-of-the-art methods, including PostgreSQL, BAO, and HybridQO. Notably, LLMOpt(S) achieves the best practical performance, striking a balance between plan quality and inference efficiency. Copyright © 2025, The Authors. All rights reserved.

关键词： Structured Query Language

NI-GDBA: Non-Intrusive Distributed Backdoor Attack Based on Adaptive Perturbation on Federated Graph Learning 34

学校读者我要写书评

暂无评论

NI-GDBA: Non-Intrusive Distributed Backdoor Attack Based on ...

34th ACM Web Conference, WWW 2025

作者： Li, Ken Shi, Bin Wei, Jiazhe Dong, Bo School of Computer Science and Technology Xi’an Jiaotong University Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi’an China School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an China School of Distance Education Xi’an Jiaotong University Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an China

ISBN: (纸本)9798400712746

Federated Graph Learning (FedGL) is an emerging Federated Learning (FL) framework that learns the graph data from various clients to train better Graph Neural Networks(GNNs) model. Owing to concerns regarding the security of such framework, numerous studies have attempted to execute backdoor attacks on FedGL, with a particular focus on distributed backdoor attacks. However, all existing methods posting distributed backdoor attack on FedGL only focus on injecting distributed backdoor triggers into the training data of each malicious client, which will cause model performance degradation on original task and is not always effective when confronted with robust federated learning defense algorithms, leading to low success rate of attack. What’s more, the backdoor signals introduced by the malicious clients may be smoothed out by other clean signals from the honest clients, which potentially undermining the performance of the attack. To address the above significant shortcomings, we propose a non-intrusive graph distributed backdoor attack(NI-GDBA) that does not require backdoor triggers to be injected in the training data. Our attack trains an adaptive perturbation trigger generator model for each malicious client to learn the natural backdoor from the GNN model downloading from the server with the malicious client’s local data. In contrast to traditional distributed backdoor attacks on FedGL via trigger injection in training data, our attack on different datasets such as Molecules and Bioinformatics have higher attack success rate, stronger persistence and stealth, and has no negative impact on the performance of the global GNN model. We also explore the robustness of NI-GDBA under different defense strategies, and based on our extensive experimental studies, we show that our attack method is robust to current federated learning defense methods, thus it is necessary to consider non-intrusive distributed backdoor attacks on FedGL as a novel threat that requires custom d

关键词： Backdoor Attacks Federated Graph Learning

HGFormer: Topology-Aware Vision Transformer with HyperGraph Learning

学校读者我要写书评

暂无评论

IEEE Transactions on Multimedia 2025年

作者： Wang, Hao Zhang, Shuo Leng, Biao Beihang University School of Computer Science and Engineering Beijing100191 China Beijing Key Laboratory of Traffic Data Analysis and Mining School of Computer Science & Technology Beijing100044 China

The computer vision community has witnessed an extensive exploration of vision transformers in the past two years. Drawing inspiration from traditional schemes, numerous works focus on introducing vision-specific inductive biases. However, the implicit modeling of permutation invariance and fully-connected interaction with individual tokens disrupts the regional context and spatial topology, further hindering higher-order modeling. This deviates from the principle of perceptual organization that emphasizes the local groups and overall topology of visual elements. Thus, we introduce the concept of hypergraph for perceptual exploration. Specifically, we propose a topology-aware vision transformer called HyperGraph Transformer (HGFormer). Firstly, we present a Center Sampling K-Nearest Neighbors (CS-KNN) algorithm for semantic guidance during hypergraph construction. Secondly, we present a topology-aware HyperGraph Attention (HGA) mechanism that integrates hypergraph topology as perceptual indications to guide the aggregation of global and unbiased information during hypergraph messaging. Using HGFormer as visual backbone, we develop an effective and unitive representation, achieving distinct and detailed scene depictions. Empirical experiments show that the proposed HGFormer achieves competitive performance compared to the recent SoTA counterparts on various visual benchmarks. Extensive ablation and visualization studies provide comprehensive explanations of our ideas and contributions. © 2025 IEEE.

关键词： Hypergraph Learning Image Classification Vision Transformer

Explicit and Implicit Consumer Electronics Information Integration via Automatic Large Language Model Construction

学校读者我要写书评

暂无评论

IEEE Transactions on Consumer Electronics 2025年

作者： Xue, Xingsi Li, Xingwang the Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology Fujian Fuzhou China the School of Physics and Electronic Information Engineering Henan Polytechnic University China

—Integrating diverse Consumer Electronics (CE) information is essential to enhance and optimize user experiences, but CE Information Integration (CEII) faces challenges arising from differences in entity descriptions. In recent years, Large Language Models (LLMs) have emerged as valuable tools for analyzing heterogeneous CE data due to their deep contextual embeddings. However, different LLMs capture various nuances and complexity levels within the CE data, and none of them can ensure their effectiveness in all heterogeneous scenarios. To address this issue, this paper proposes an automatic LLM Construction for CEII, which can determine both explicit and implicit CE entity mappings. First, a hybrid Genetic Programming is developed to determine explicit entity mappings using a new individual representation, an innovative fitness function, and a Probability-based Incremental Learning algorithm. Then, a novel Rough Set based matching method is presented to efficiently induce implicit CE entity mappings, which uses the entity feature relevance metric and the dominance-based matching approach. The experiment tests the performance of our approach with the OAEI’s KG and common KG datasets, and the practical CE datasets in the smart home context. The experimental results show the effectiveness of the developed approach, which significantly outperforms the state-of-the-art entity matching methods. © 2025 IEEE.

关键词： Genetic programming

Effective Instruction Parsing Plugin for Complex Logical Query Answering on knowledge Graphs 34

学校读者我要写书评

暂无评论

Effective Instruction Parsing Plugin for Complex Logical Que...

34th ACM Web Conference, WWW 2025

作者： Zhuo, Xingrui Wang, Jiapu Wu, Gongqing Pan, Shirui Wu, Xindong Hefei University of Technology Anhui Hefei China Beijing University of Technology Beijing China Griffith University Gold Coast QLD Australia The Key Laboratory of Knowledge Engineering with Big Data (the Ministry of Education of China) School of Computer Science and Information Engineering Hefei University of Technology China

ISBN: (纸本)9798400712746

knowledge Graph Query Embedding (KGQE) aims to embed First-Order Logic (FOL) queries in a low-dimensional KG space for complex reasoning over incomplete KGs. To enhance the generalization of KGQE models, recent studies integrate various external information (such as entity types and relation context) to better capture the logical semantics of FOL queries. The whole process is commonly referred to as Query Pattern Learning (QPL). However, current QPL methods typically suffer from the pattern-entity alignment bias problem, leading to the learned defective query patterns limiting KGQE models’ performance. To address this problem, we propose an effective Query Instruction Parsing Plugin (QIPP) that leverages the context awareness of Pre-trained Language Models (PLMs) to capture latent query patterns from code-like query instructions. Unlike the external information introduced by previous QPL methods, we first propose code-like instructions to express FOL queries in an alternative format. This format utilizes textual variables and nested tuples to convey the logical semantics within FOL queries, serving as raw materials for a PLM-based instruction encoder to obtain complete query patterns. Building on this, we design a query-guided instruction decoder to adapt query patterns to KGQE models. To further enhance QIPP’s effectiveness across various KGQE models, we propose a query pattern injection mechanism based on compressed optimization boundaries and an adaptive normalization component, allowing KGQE models to utilize query patterns more efficiently. Extensive experiments demonstrate that our plug- and-play method¹ improves the performance of eight basic KGQE models and outperforms two state-of-the-art QPL methods. © 2025 Copyright held by the owner/author(s).

关键词： Complex Query Answering knowledge Graph Pre-trained Language Model Query Pattern Learning

LLMIdxAdvis: Resource-Efficient Index Advisor Utilizing Large Language Model

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhao, Xinxin Li, Haoyang Zhang, Jing Huang, Xinmei Zhang, Tieying Chen, Jianjun Shi, Rui Li, Cuiping Chen, Hong School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering MOE China Engineering Research Center of Database and Business Intelligence MOE China ByteDance China

Index recommendation is essential for improving query performance in database management systems (DBMSs) through creating an optimal set of indexes under specific constraints. Traditional methods, such as heuristic and learning-based approaches, are effective but face challenges like lengthy recommendation time, resource-intensive training, and poor generalization across different workloads and database schemas. To address these issues, we propose LLMIdxAdvis, a resource-efficient index advisor that uses large language models (LLMs) without extensive fine-tuning. LLMIdxAdvis frames index recommendation as a sequence-to-sequence task, taking target workload, storage constraint, and corresponding database environment as input, and directly outputting recommended indexes. It constructs a high-quality demonstration pool offline, using GPT-4-Turbo to synthesize diverse SQL queries and applying integrated heuristic methods to collect both default and refined labels. During recommendation, these demonstrations are ranked to inject database expertise via in-context learning. Additionally, LLMIdxAdvis extracts workload features involving specific column statistical information to strengthen LLM’s understanding, and introduces a novel inference scaling strategy combining vertical scaling (via "Index-Guided Major Voting" and Best-of-N) and horizontal scaling (through iterative "self-optimization" with database feedback) to enhance reliability. Experiments on 3 OLAP and 2 real-world benchmarks reveal that LLMIdxAdvis delivers competitive index recommendation with reduced runtime, and generalizes effectively across different workloads and database schemas. Copyright © 2025, The Authors. All rights reserved.

关键词： Query languages

OmniSQL: Synthesizing High-quality Text-to-SQL data at Scale

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Haoyang Wu, Shang Zhang, Xiaokang Huang, Xinmei Zhang, Jing Jiang, Fuxin Wang, Shuai Zhang, Tieying Chen, Jianjun Shi, Rui Chen, Hong Li, Cuiping Engineering Research Center of Database and Business Intelligence MOE China School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering MOE China ByteDance Inc China

Text-to-SQL, the task of translating natural language questions into SQL queries, plays a crucial role in enabling non-experts to interact with databases. While recent advancements in large language models (LLMs) have significantly enhanced text-to-SQL performance, existing approaches face notable limitations in real-world text-to-SQL applications. Prompting-based methods often depend on closed-source LLMs, which are expensive, raise privacy concerns, and lack customization. Fine-tuning-based methods, on the other hand, suffer from poor generalizability due to the limited coverage of publicly available training data. To overcome these challenges, we propose a novel and scalable text-to-SQL data synthesis framework for automatically synthesizing large-scale, high-quality, and diverse datasets without extensive human intervention. Using this framework, we introduce SynSQL-2.5M, the first million-scale text-to-SQL dataset, containing 2.5 million samples spanning over 16,000 synthetic databases. Each sample includes a database, SQL query, natural language question, and chain-of-thought (CoT) solution. Leveraging SynSQL-2.5M, we develop OmniSQL, a powerful open-source text-to-SQL model available in three sizes: 7B, 14B, and 32B. Extensive evaluations across nine datasets demonstrate that OmniSQL achieves state-of-the-art performance, matching or surpassing leading closed-source and open-source LLMs, including GPT-4o and DeepSeek-V3, despite its smaller size. We release all code, datasets, and models to support further research. Copyright © 2025, The Authors. All rights reserved.

关键词： Structured Query Language

Joint Double Auction-Based Channel Selection in Wireless Monitoring Networks

学校读者我要写书评

暂无评论

IEEE Transactions on Network and Service Management 2025年

作者： Xia, Na Chen, Lei Li, Meng Yin, Yutao Zhang, Ke Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology Key Laboratory of Knowledge Engineering with Big Data China State Grid Electric Power Research Institute Co. Ltd Hefei230088 China

In wireless networks, utilizing sniffers for fault analysis, traffic traceback, and resource optimization is a crucial task. However, existing centralized algorithms cannot be applied to high-density wireless networks. Therefore, distributed optimization of channel selection to maximize the monitoring rate of sensors in Wireless Monitoring Networks (WMNs) is a challenge. This paper proposes a joint double auction-based distributed channel selection algorithm (J2A-CS) to maximize overall quality of monitoring (QoM). First, sniffers are redundantly deployed in WMNs, and an initial channel allocation strategy is formulated. Subsequently, sniffers collectively act as buyers and sellers at different stages. Finally, buyers bid asynchronously, and sellers settle synchronously to maximize the seller's marginal revenue and update the channel selection scheme. As a distributed channel selection algorithm, J2A-CS addresses the highest overall QoM issue in WMNs, demonstrating high scalability and fault tolerance. Simulation results show that J2A-CS significantly improves QoM compared to existing distributed algorithms and outperforms centralized algorithms in high-density scenarios. © 2004-2012 IEEE.

关键词： Sales

LoRS: Efficient Low-Rank Adaptation for Sparse Large Language Model

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hu, Yuxuan Zhang, Jing Chen, Xiaodong Zhao, Zhe Li, Cuiping Chen, Hong School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering Beijing China Engineering Research Center of Database and Business Intelligence Beijing China Tencent AI Lab Beijing China

Existing low-rank adaptation (LoRA) methods face challenges on sparse large language models (LLMs) due to the inability to maintain sparsity. Recent works introduced methods that maintain sparsity by augmenting LoRA techniques with additional masking mechanisms. Despite these successes, such approaches suffer from an increased memory and computation overhead, which affects efficiency of LoRA methods. In response to this limitation, we introduce LoRS, an innovative method designed to achieve both memory and computation efficiency when fine-tuning sparse LLMs. To mitigate the substantial memory and computation demands associated with preserving sparsity, our approach incorporates strategies of weight recompute and computational graph rearrangement. In addition, we also improve the effectiveness of LoRS through better adapter initialization. These innovations lead to a notable reduction in memory and computation consumption during the fine-tuning phase, all while achieving performance levels that outperform existing LoRA approaches. © 2025, CC BY.

关键词： Problem oriented languages