检索结果-内蒙古大学图书馆

MoC: Mixtures of Text Chunking Learners for Retrieval-Augmented Generation System

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhao, Jihao Ji, Zhiyuan Fan, Zhaoxin Wang, Hanyu Niu, Simin Tang, Bo Xiong, Feiyu Li, Zhiyu School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China

Retrieval-Augmented Generation (RAG), while serving as a viable complement to large language models (LLMs), often overlooks the crucial aspect of text chunking within its pipeline. This paper initially introduces a dual-metric evaluation method, comprising Boundary Clarity and Chunk Stickiness, to enable the direct quantification of chunking quality. Leveraging this assessment method, we highlight the inherent limitations of traditional and semantic chunking in handling complex contextual nuances, thereby substantiating the necessity of integrating LLMs into chunking process. To address the inherent trade-off between computational efficiency and chunking precision in LLM-based approaches, we devise the granularity-aware Mixture-of-Chunkers (MoC) framework, which consists of a three-stage processing mechanism. Notably, our objective is to guide the chunker towards generating a structured list of chunking regular expressions, which are subsequently employed to extract chunks from the original text. Extensive experiments demonstrate that both our proposed metrics and the MoC framework effectively settle challenges of the chunking task, revealing the chunking kernel while enhancing the performance of the RAG system 1. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

When Sparse Graph Representation Learning Falls into Domain Shift: Feature Augmentation for Cross-Domain Graph Meta-Learning

学校读者我要写书评

暂无评论

When Sparse Graph Representation Learning Falls into Domain ...

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Niu, Simin Liang, Xun Zhang, Sensen Li, Zhiyu Zhang, Xuan Bo, Wu Wang, Hanyu Song, Shichao Wang, Mengwei Yang, Jiawei School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China Guanghua School of Management Peking University Beijing China Xiangjiang Laboratory Central South University Hunan China

ISBN: (纸本)9798350368741

Graph Meta-learning methods have improved the performance of few-shot node classification by means of applying meta-learning to the data in non-Euclidean domains. However, most works focus on adopting a single domain, ignoring the fact that tasks in various domains may be distinct, which can cause overfitting problems and thus limit generalizability. To tackle this challenge, we propose a novel Graph Meta-learning framework called Feature-Enhanced Cross-domain Graph Meta-learning that consists of two crucial modules: 1) A feature information extraction module that aims to capture discriminative node importance and simulate various node feature distributions under distinct domains;2) A heterogeneous graph encoder module that leverages the enhanced node features and topological information to generate task-specific node embeddings with simple fine-tuning. Moreover, we meta-learn the parameters involved to ensure the generalizability in the unseen domains. Results show that our method markedly outperforms the existing state-of-the-art methods. © 2025 IEEE.

关键词： Domain Generalization Few-shot Learning Heterogeneous Graph Neural Networks Meta-learning

Retrieval-Augmented Multilingual Citation Generation

学校读者我要写书评

暂无评论

Retrieval-Augmented Multilingual Citation Generation

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Liang, Xun Niu, Simin Zhang, Sensen Li, Zhiyu Zhang, Xuan Wu, Bo Xiong, Feiyu Tang, Bo Wang, Hanyu Song, Shichao Wang, Mengwei Yang, Jiawei School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China Guanghua School of Management Peking University Beijing China Xiangjiang Laboratory Central South University Hunan China

ISBN: (纸本)9798350368741

Retrieval-augmented citation generation (RACG) helps users trust the large language model output by retrieving evidence from reliable sources. However, most current RACG research focuses on single-language tasks, particularly in English, and overlooks the need for cross-lingual evidence retrieval and utilization in real-world applications. To address this issue, we introduce a plug-and-play Retrieval-Augmented Multilingual Citation Generation method (RAMCG) which uses a multilingual retriever to identify relevant evidence from a multilingual knowledge base. The evidence is then combined with the query and processed by a multilingual citation generator. The result is citations that are both accurate and comprehensive. Experiments show that RAMCG outperforms baseline methods in multilingual citation generation and is well-suited for practical use. © 2025 IEEE.

关键词： Applications of Generative AI Citation Generation Retrieval-Augmented Generation

HopRAG: Multi-Hop Reasoning for Logic-Aware Retrieval-Augmented Generation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Hao Wang, Zhengren Chen, Xi Li, Zhiyu Xiong, Feiyu Yu, Qinhan Zhang, Wentao Peking University China Center for LLM Institute for Advanced Algorithms Research Shanghai China Huazhong University of Science and Technology China

Retrieval-Augmented Generation (RAG) systems often struggle with imperfect retrieval, as traditional retrievers focus on lexical or semantic similarity rather than logical relevance. To address this, we propose HopRAG, a novel RAG framework that augments retrieval with logical reasoning through graph-structured knowledge exploration. During indexing, HopRAG constructs a passage graph, with text chunks as vertices and logical connections established via LLM-generated pseudo-queries as edges. During retrieval, it employs a retrieve-reason-prune mechanism: starting with lexically or semantically similar passages, the system explores multi-hop neighbors guided by pseudo-queries and LLM reasoning to identify truly relevant ones. Extensive experiments demonstrate HopRAG’s superiority, achieving 76.78% higher answer accuracy and 65.07% improved retrieval F1 score compared to conventional methods. The repository is available at https://***/LIU-Hao-2002/HopRAG. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

SurveyX: Academic Survey Automation via Large Language Models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liang, Xun Yang, Jiawei Wang, Yezhaohui Tang, Chen Zheng, Zifan Song, Shichao Niu, Simin Wang, Hanyu Tang, Bo Xiong, Feiyu Mao, Keming Li, Zhiyu Renmin University of China Beijing China Northeastern University Shenyang China Institute for Advanced Algorithms Research Shanghai China The University of Sydney Sydney Australia

Large Language Models (LLMs) have demonstrated exceptional comprehension capabilities and a vast knowledge base, suggesting that LLMs can serve as efficient tools for automated survey generation. However, recent research related to automated survey generation remains constrained by some critical limitations like finite context window, lack of in-depth content discussion, and absence of systematic evaluation frameworks. Inspired by human writing processes, we propose SurveyX, an efficient and organized system for automated survey generation that decomposes the survey composing process into two phases: the Preparation and Generation phases. By innovatively introducing online reference retrieval, a pre-processing method called AttributeTree, and a re-polishing process, SurveyX significantly enhances the efficacy of survey composition. Experimental evaluation results show that SurveyX outperforms existing automated survey generation systems in content quality (0.259 improvement) and citation quality (1.76 enhancement), approaching human expert performance across multiple evaluation dimensions. Examples of surveys generated by SurveyX are available on our project website Copyright © 2025, The Authors. All rights reserved.

关键词： Modeling languages

SafeRAG: Benchmarking Security in Retrieval-Augmented Generation of Large Language Model

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liang, Xun Niu, Simin Li, Zhiyu Zhang, Sensen Wang, Hanyu Xiong, Feiyu Fan, Jason Zhaoxin Tang, Bo Song, Shichao Wang, Mengwei Yang, Jiawei School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing School of Artificial Intelligence Beihang University Beijing China

The indexing-retrieval-generation paradigm of retrieval-augmented generation (RAG) has been highly successful in solving knowledge-intensive tasks by integrating external knowledge into large language models (LLMs). However, the incorporation of external and unverified knowledge increases the vulnerability of LLMs because attackers can perform attack tasks by manipulating knowledge. In this paper, we introduce a benchmark named SafeRAG designed to evaluate the RAG security. First, we classify attack tasks into silver noise, inter-context conflict, soft ad, and white Denial-of-Service. Next, we construct RAG security evaluation dataset (i.e., SafeRAG dataset) primarily manually for each task. We then utilize the SafeRAG dataset to simulate various attack scenarios that RAG may encounter. Experiments conducted on 14 representative RAG components demonstrate that RAG exhibits significant vulnerability to all attack tasks and even the most apparent attack task can easily bypass existing retrievers, filters, or advanced LLMs, resulting in the degradation of RAG service quality. Code is available at: https://***/IAAR-Shanghai/SafeRAG. © 2025, CC BY.

关键词： HTTP

When Sparse Graph Representation Learning Falls into Domain Shift: Feature Augmentation for Cross-Domain Graph Meta-Learning

学校读者我要写书评

暂无评论

When Sparse Graph Representation Learning Falls into Domain ...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Simin Niu Xun Liang Sensen Zhang Zhiyu Li Xuan Zhang Wu Bo Hanyu Wang Shichao Song Mengwei Wang Jiawei Yang School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China Guanghua School of Management Peking University Beijing China Xiangjiang Laboratory Central South University Hunan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Graph Meta-learning methods have improved the performance of few-shot node classification by means of applying meta-learning to the data in non-Euclidean domains. However, most works focus on adopting a single domain, ignoring the fact that tasks in various domains may be distinct, which can cause overfitting problems and thus limit generalizability. To tackle this challenge, we propose a novel Graph Meta-learning framework called Feature-Enhanced Cross-domain Graph Meta-learning that consists of two crucial modules: 1) A feature information extraction module that aims to capture discriminative node importance and simulate various node feature distributions under distinct domains; 2) A heterogeneous graph encoder module that leverages the enhanced node features and topological information to generate task-specific node embeddings with simple fine-tuning. Moreover, we meta-learn the parameters involved to ensure the generalizability in the unseen domains. Results show that our method markedly outperforms the existing state-of-the-art methods.

关键词： Metalearning Representation learning Signal processing Feature extraction Information retrieval Acoustics Data mining Speech processing Overfitting

Retrieval-Augmented Multilingual Citation Generation

学校读者我要写书评

暂无评论

Retrieval-Augmented Multilingual Citation Generation

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xun Liang Simin Niu Sensen Zhang Zhiyu Li Xuan Zhang Bo Wu Feiyu Xiong Bo Tang Hanyu Wang Shichao Song Mengwei Wang Jiawei Yang School of Information Renmin University of China Beijing China Institute for Advanced Algorithms Research Shanghai China Guanghua School of Management Peking University Beijing China Xiangjiang Laboratory Central South University Hunan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

关键词： Measurement Generative AI Large language models Knowledge based systems Signal processing Generators Acoustics Multilingual Reliability Speech processing

GRAPHMOE: Amplifying Cognitive Depth of Mixture-of-Experts Network via Introducing Self-Rethinking Mechanism

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Tang, Chen Lv, Bo Zheng, Zifan Yang, Bohao Zhao, Kun Liao, Ning Wang, Xiaoxing Xiong, Feiyu Li, Zhiyu Liu, Nayu Jiang, Jingchi Institute for Advanced Algorithms Research Shanghai China Institute of Computing Technology Chinese Academy of Sciences China The University of Manchester United Kingdom The University of Pittsburgh United States National Key Laboratory of Smart Farm Technologies and Systems Harbin Institute of Technology China University of Sydney Australia

Traditional Mixture-of-Experts (MoE) networks benefit from utilizing multiple smaller expert models as opposed to a single large network. However, these experts typically operate independently, leaving a question open about whether interconnecting these models could enhance the performance of MoE networks. In response, we introduce GRAPHMOE, a novel method aimed at augmenting the cognitive depth of language models via a self-rethinking mechanism constructed on Pseudo Graph MoE networks. GRAPHMOE employs a recurrent routing strategy to simulate iterative thinking steps, thereby facilitating the flow of information among expert nodes. We implement the GRAPHMOE architecture using Low-Rank Adaptation techniques (LoRA) and conduct extensive experiments on various benchmark datasets. The experimental results reveal that GRAPHMOE outperforms other LoRA based models, achieving state-of-the-art (SOTA) performance. Additionally, this study explores a novel recurrent routing strategy that may inspire further advancements in enhancing the reasoning capabilities of language models. Copyright © 2025, The Authors. All rights reserved.

关键词： Network architecture