检索结果-内蒙古大学图书馆

LLM-Driven Knowledge Injection Advances Zero-Shot and Cross-Target Stance Detection

学校读者我要写书评

暂无评论

LLM-Driven Knowledge Injection Advances Zero-Shot and Cross-...

2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2024

作者： Zhang, Zhao Li, Yiming Zhang, Jin Xu, Hui CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China Beijing Key Laboratory of Mobile Computing and Pervasive Device Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China

Stance detection aims at inferring an author’s attitude towards a specific target in a text. Prior methods mainly consider target-related background information for a better understanding of targets while neglecting the accompanying input texts. In this study, we propose to prompt Large Language Models (LLMs) to explicitly extract the relationship between paired text and target as contextual knowledge. We then inject such LLM-driven knowledge into a generation model BART to exploit the rich contexts and semantics. Moreover, to further enhance the decoding capability of BART, a novel prototypical contrastive scheme is designed to align input contents with stance labels. Our experimental results demonstrate the state-of-the-art performance across several publicly available datasets, showcasing effectiveness in both zero-shot and cross-target stance detection scenarios. We publicly release our code to facilitate future research. © 2024 Association for Computational Linguistics.

关键词： Zero-shot learning

Spherical paragraph model

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zhang, Ruqing Guo, Jiafeng Lan, Yanyan Xu, Jun Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China

Representing texts as fixed-length vectors is central to many language processing tasks. Most traditional methods build text representations based on the simple Bag-of-Words (BoW) representation, which loses the rich semantic relations between words. Recent advances in natural language processing have shown that semantically meaningful representations of words can be efficiently acquired by distributed models, making it possible to build text representations based on a better foundation called the Bag-of-Word-Embedding (BoWE) representation. However, existing text representation methods using BoWE often lack sound probabilistic foundations or cannot well capture the semantic relatedness encoded in word vectors. To address these problems, we introduce the Spherical Paragraph Model (SPM), a probabilistic generative model based on BoWE, for text representation. SPM has good probabilistic interpretability and can fully leverage the rich semantics of words, the word co-occurrence information as well as the corpus-wide information to help the representation learning of texts. Experimental results on topical classification and sentiment analysis demonstrate that SPM can achieve new state-of-the-art performances on several benchmark datasets. Copyright © 2017, The Authors. All rights reserved.

关键词： Sentiment analysis

MatchZoo: A toolkit for deep text matching

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Fan, Yixing Pang, Liang Hou, JianPeng Guo, Jiafeng Lan, Yanyan Cheng, Xueqi Cas Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China

In recent years, deep neural models have been widely adopted for text matching tasks, such as question answering and information retrieval, showing improved performance as compared with previous methods. In this paper, we introduce the MatchZoo toolkit that aims to facilitate the designing, comparing and sharing of deep text matching models. Specifically, the toolkit provides a unified data preparation module for different text matching problems, a flexible layer-based model construction process, and a variety of training objectives and evaluation metrics. In addition, the toolkit has implemented two schools of representative deep text matching models, namely representation-focused models and interaction-focused models. Finally, users can easily modify existing models, create and share their own models for text matching in MatchZoo. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep learning

Classifier Guidance Enhances Diffusion-Based Adversarial Purification by Preserving Predictive Information 27

学校读者我要写书评

暂无评论

Classifier Guidance Enhances Diffusion-Based Adversarial Pur...

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Zhang, Mingkun Li, Jianing Chen, Wei Guo, Jiafeng Cheng, Xueqi CAS Key Laboratory of AI Safety Institute of Computing Technology Chinese Academy of Sciences Beijing China Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781643685489

Adversarial purification is one of the promising approaches to defend neural networks against adversarial attacks. Recently, methods utilizing diffusion probabilistic models have achieved great success for adversarial purification in image classification tasks. However, such methods fall into the dilemma of balancing the needs for noise removal and information preservation. This paper points out that existing adversarial purification methods based on diffusion models gradually lose sample information during the core denoising process, causing occasional label shift in subsequent classification tasks. As a remedy, we suggest to suppress such information loss by introducing guidance from the classifier confidence. Specifically, we propose Classifier-cOnfidence gUided Purification (COUP) algorithm, which purifies adversarial examples while keeping away from the classifier decision boundary. Experimental results show that COUP can achieve better adversarial robustness under strong attack methods. © 2024 The Authors.

关键词： Generative adversarial networks

An enhanced genetic algorithm for computation task offloading in MEC scenario

学校读者我要写书评

暂无评论

International Journal of Wireless and Mobile Computing 2023年第2期25卷 118-127页

作者： Zhao, Jiacheng Li, Wenzao Liu, Hantao Yu, Peizhen Li, Hanyun Wen, Zhan School of Communication Engineering Chengdu University of Information Technology Sichuan Chengdu China Chengdu University of Information Technology Sichuan Chengdu China Network and Data Security Key Lab of Sichuan Pro. University of Electronic Science and Technology of China Sichuan Chengdu China Educational Informationisation and Big Data Centre Education Department of Sichuan Province Sichuan Chengdu China

The explosive growth of Internet of Things (IoT) and 5G communication technologies has driven the increasing computing demands for wireless devices. Mobile edge computing in the 5G scenario is a promising solution for energy-efficient and low latency applications. However, due to limited bandwidth, the selection of appropriate computing tasks greatly affects the user experience and system performance. Under the wireless bandwidth constraint, the reasonable choice of offloading objects is an NP-hard problem. The genetic algorithm has a great ability to solve this problem, but the performance of the algorithm varies with different scenarios. This paper proposes a task offloading strategy based on an enhanced genetic algorithm for small-scale computing tasks with an ultra-dense terminal distribution. Numerical experiments show that the convergence speed and optimisation effect of the enhanced genetic algorithm are significantly improved compared to the conventional genetic algorithm. © 2023 Inderscience Enterprises Ltd.

关键词： Genetic algorithms

A tree search algorithm for sequence labeling

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Lao, Yadi Xu, Jun Lan, Yanyan Guo, Jiafeng Gao, Sheng Cheng, Xueqi Guo, Jun Beijing University of Posts and Telecommunications CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Science

In this paper we propose a novel reinforcement learning based model for sequence tagging, referred to as MM-Tag. Inspired by the success and methodology of the AlphaGo Zero, MM-Tag formalizes the problem of sequence tagging with a Monte Carlo tree search (MCTS) enhanced Markov decision process (MDP) model, in which the time steps correspond to the positions of words in a sentence from le. to right, and each action corresponds to assign a tag to a word. Two long short-term memory networks (LSTM) are used to summarize the past tag assignments and words in the sentence. Based on the outputs of LSTMs, the policy for guiding the tag assignment and the value for predicting the whole tagging accuracy of thewhole sentence are produced.e policy and value are then strengthened with MCTS, which takes the produced raw policy and value as inputs, simulates and evaluates the possible tag assignments at the subsequent positions, and outputs a *** search policy for assigning tags. A reinforcement learning algorithm is proposed to train the model parameters. Our work is the first to apply the MCTS enhanced MDP model to the sequence tagging task. We show that MM-Tag can accurately predict the tags thanks to the exploratory decision making mechanism introduced by MCTS. Experimental results show based on a chunking benchmark showed that MM-Tag outperformed the state-of-the-art sequence tagging baselines including CRF and CRF with LSTM. Copyright © 2018, The Authors. All rights reserved.

关键词： Long short-term memory

Quantum-to-quantum Bernoulli factory problem

学校读者我要写书评

暂无评论

Physical Review A 2018年第3期97卷 032303-032303页

作者： Jiaqing Jiang Jialin Zhang Xiaoming Sun CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China University of Chinese Academy of Sciences Beijing 100049 China

Given a coin with unknown bias p∈[0,1], can we exactly simulate another coin with bias f(p)? The exact set of simulable functions has been well characterized 20 years ago. In this paper, we ask the quantum counterpart of this question: Given the quantum coin |p〉=p|0〉+1−p|1〉, can we exactly simulate another quantum coin |f(p)〉=f(p)|0〉+1−f(p)|1〉? We give the full characterization of simulable quantum state k0(p)|0〉+k1(p)|1〉 from quantum coin |p〉=p|0〉+1−p|1〉, and present an algorithm to transform it. Surprisingly, we show that simulable sets in the quantum-to-quantum case and classical-to-classical case have no inclusion relationship with each other.

关键词： Quantum algorithms Quantum computation Quantum simulation

Controlling Risk of Retrieval-augmented Generation: A Counterfactual Prompting Framework

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Lu Zhang, Ruqing Guo, Jiafeng Fan, Yixing Cheng, Xueqi CAS Key Lab of Network Data Science and Technology ICT CAS Beijing China University of Chinese Academy of Sciences Beijing China

Retrieval-augmented generation (RAG) has emerged as a popular solution to mitigate the hallucination issues of large language models. However, existing studies on RAG seldom address the issue of predictive uncertainty, i.e., how likely it is that a RAG model’s prediction is incorrect, resulting in uncontrollable risks in real-world applications. In this work, we emphasize the importance of risk control, ensuring that RAG models proactively refuse to answer questions with low confidence. Our research identifies two critical latent factors affecting RAG’s confidence in its predictions: the quality of the retrieved results and the manner in which these results are utilized. To guide RAG models in assessing their own confidence based on these two latent factors, we develop a counterfactual prompting framework that induces the models to alter these factors and analyzes the effect on their answers. We also introduce a benchmarking procedure to collect answers with the option to abstain, facilitating a series of experiments. For evaluation, we introduce several risk-related metrics and the experimental results demonstrate the effectiveness of our approach. Our code and benchmark dataset are available at https://***/ictbigdatalab/RC-RAG. Copyright © 2024, The Authors. All rights reserved.

关键词： Risk assessment

Self-Learning and Embedding Based Entity Alignment

学校读者我要写书评

暂无评论

Self-Learning and Embedding Based Entity Alignment

IEEE International Conference on Big Knowledge (ICBK)

作者： Saiping Guan Xiaolong Jin Yantao Jia Yuanzhuo Wang Huawei Shen Xueqi Cheng CAS Key Laboratory of Network Data Science and Technology University of Chinese Academy of Sciences

ISBN: (纸本)9781538631218

Entity alignment aims to identify semantical matchings between entities from different groups. Traditional methods (e.g., attribute comparison based methods, clustering based methods, and active learning methods) are usually supervised by labelled data as prior knowledge. Since it is not trivial to label data for training, researchers have recently turned to unsupervised methods, and have thus developed similarity based methods, probabilistic methods, hierarchical graph model based methods, etc. As an important part of a knowledge graph, entities contain rich semantical information that can be well learned by knowledge graph embedding in a low-dimensional vector space. However, existing methods for entity alignment have paid little attention to knowledge graph embedding. In this paper, we propose a Self-learning and Embedding based method for Entity Alignment, thus called SEEA, to iteratively find semantically matched entity pairs, which makes full use of semantical information contained in the attributes of entities. Experiments on two realistic datasets and comparison with the baselines validate the effectiveness and merits of the proposed method.

关键词： Indexes Motion pictures Probabilistic logic Ontologies Training data models