检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Li, Zhixuan Yoon, Hyunse Lee, Sanghoon Lin, Weisi College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical Electronic Engineering Yonsei University Korea Republic of

Amodal segmentation aims to infer the complete shape of occluded objects, even when the occluded region’s appearance is unavailable. However, current amodal segmentation methods lack the capability to interact with users through text input and struggle to understand or reason about implicit and complex purposes. While methods like LISA integrate multi-modal large language models (LLMs) with segmentation for reasoning tasks, they are limited to predicting only visible object regions and face challenges in handling complex occlusion scenarios. To address these limitations, we propose a novel task named amodal reasoning segmentation, aiming to predict the complete amodal shape of occluded objects while providing answers with elaborations based on user text input. We develop a generalizable dataset generation pipeline and introduce a new dataset focusing on daily life scenarios, encompassing diverse real-world occlusions. Furthermore, we present AURA (Amodal Understanding and Reasoning Assistant), a novel model with advanced global and spatial-level designs specifically tailored to handle complex occlusions. Extensive experiments validate AURA’s effectiveness on the proposed dataset. The code, model, and dataset will be publicly released. Copyright © 2025, The Authors. All rights reserved.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

Predictive Analysis of Consumer Behavior Using Supervised Learning Techniques 2

Predictive Analysis of Consumer Behavior Using Supervised Le...

引用

2nd International Conference Computational and Characterization Techniques in Engineering and sciences, IC3TES 2024

作者： Rekha, K. Sashi Amutha, T. Usharani, R. Pushparani, S. Sivagami, V.M. Saveetha Institute of Medical and Technical Sciences Simats Engineering Department of Computer Science and Engineering Tamil Nadu Chennai India Care College of Engineering Department of Artificial Intelligence & Data Science Tamil Nadu Trichy India Srm Institute of Science and Technology Faculty of Engineering and Technology Department of Computational Intelligence School of Computing Chennai India Meenakshi College of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai India Sri Venkateswara College of Engineering Department of Information Technology Tamil Nadu Chennai India

ISBN: (纸本)9798350364699

The study presented here consists of an entire framework for prediction of consumers behavior through the machine training of supervised learning methods. The strategy involves the following: data collection, preprocessing, exploration, feature selection, model selection, training, evaluation, interpretation and deployment. The systematic process is tailored to clear the pathway for combining data about customer taste, sophisticated habits, and purchasing trends. Primary components include data preprocessing, training of machine learning models, feature selection, and ethical issues regarding data privacy and transparency. The outcomes of the implementation are a high accuracy in consumers clustering, efficient classification of customers' behavior, as well as unveiling a bunch of crucial predictors that affect the customers mood and behavior. As a result, the report stresses the necessity of integrated monitoring and redirection strategies to realize the potential of the created models. The idea of predictive analysis as an evolving tool that enables marketers to tailor their campaigns more closely, and improve the overall customer experience therefore also receives attention. The proposed system methodology constitutes a solid process framework for business defined to aim at using predictive analytics to gain competitive edge in the current dynamic marketplace. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

When Words Outperform Vision: VLMs Can Self-Improve Via Text-Only Training For Human-Centered Decision Making

arXiv

引用

arXiv 2025年

作者： Hu, Zhe Li, Jing Yin, Yu Department of Computing The Hong Kong Polytechnic University Hong Kong Research Centre for Data Science & Artificial Intelligence Department of Computer and Data Sciences Case Western Reserve University United States

Embodied decision-making is fundamental for AI agents operating in real-world environments. While Visual Language Models (VLMs) have advanced this capability, they still struggle with complex decisions, particularly in human-centered situations that require deep reasoning about human needs and values. In this study, we systematically evaluate open-sourced VLMs on multimodal human-centered decision-making tasks. We find that LLMs receiving only textual descriptions unexpectedly outperform their VLM counterparts of similar scale that process actual images, suggesting that visual alignment may hinder VLM abilities. To address this challenge, we propose a novel text-only training approach with synthesized textual data. This method strengthens VLMs’ language components and transfers the learned abilities to multimodal inference, eliminating the need for expensive image-text paired data. Furthermore, we show that VLMs can achieve substantial performance gains through self-improvement, using training data generated by their LLM counterparts rather than relying on larger teacher models like GPT-4. Our findings establish a more efficient and scalable approach to enhancing VLMs’ human-centered decision-making capabilities, opening new avenues for optimizing VLMs through self-improvement mechanisms. © 2025, CC BY-NC-SA.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

An efficient information retrieval and tracking algorithm for multiple typesetting styles of the same paper 7

An efficient information retrieval and tracking algorithm fo...

引用

7th International Conference on data science and Information Technology, DSIT 2024

作者： Liu, Yingying Li, Yuan Tang, Jiyue Ji, Tong Xu, Xiaoyu Luo, Yuxin Lin, Yifeng Yang, Yuer Wu, Yubing Jinan University Library Jinan University Guangzhou China Jinan University College of Information Science and Technology Guangzhou China Nanyang Technological University College of Computing and Data Science Nanyang Ave Singapore Jinan University College of Cyber Security Guangzhou China The University of Hong Kong Department of Computer Science Hong Kong Hong Kong

ISBN: (纸本)9798350384093

Academic paper publishers like IEEE, Elsevier, and Springer often require manuscripts to be typeset according to their own templates before they are submitted. Most journals and conferences make final and binding decisions, where the decisions for the last time cannot be discussed or changed again in the future unless a new manuscript or a revised manuscript whose revision rate exceeds a certain large percentage is submitted. This makes authors have to change the typesetting of their rejected manuscripts on most occasions before transferring their manuscripts to a more suitable place. Unfortunately, when either the typesetting style or the manuscript content needs to be changed or updated, it is very inconvenient and time-consuming for the traditional typesetting method to re-format the manuscript since it is cumbersome to manage multiple versions of LaTeX folders. Although online platforms like Overleaf support multiplayer online collaboration, they cannot fundamentally solve the problem of generating several manuscripts of different typesetting styles with the same content by updating the manuscript content only once. To make it more flexible to change the typesetting styles, we aim to provide a novel typesetting method that supports multiple typesetting styles of the same manuscript at the same time. The manuscript content and the typesetting controlling parts are entirely separated. Meanwhile, the grammar detection and structure tracking of LaTeX source codes are presented in a way that corresponds to the new typesetting method, which can provide more appropriate results. Compared with the traditional method, the proposed typesetting method can efficiently generate multiple typesetting styles for the same manuscript by reducing the time complexity from O(mn) to O(m) on m updates with n typesetting templates, whereas the proposed grammar detection and structure tracking algorithm can be performed more efficiently and correctly. © 2024 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

Disentangled latent factors for muti-cause treatment effect estimation 6

Disentangled latent factors for muti-cause treatment effect ...

引用

2024 6th International Conference on Artificial Intelligence Technologies and Applications, ICAITA 2024

作者： Zhao, Zhuoyue Wang, Haotian Xu, Liyang Wang, Qian Tang, Yuhua Department of Intelligent Data Science National University of Defense Technology Changsha410073 China Intelligent Game and Decision Lab. Academy of Military Science Beijing100071 China Institute for Quantum Information State Key Laboratory of High-Performance Computing College of Computer Science and Technology National University of Defense Technology Changsha410073 China

Existing methods estimate treatment effects from observational data and assume that covariates are all confounders. However, observed covariates may not directly represent confounding variables that influence both treatment and outcome. They always include variables that only affect the treatment or the outcome. In addition, for multi-dimensional binary treatments, disentangled methods are mainly designed for binary variables and ignore the impact of multi-cause treatment variables on the inference of latent factors. To address these two issues, based on variable decomposition and proxy inference, we propose the Disentangled Latent Factors for Multi-cause Treatment Estimation (DEMTE) algorithm. It utilizes an identifiable autoencoder to infer and disentangle latent factors based on the joint distribution of variables in observational data. DEMTE evaluates the treatment effect on the disentangled factors. Synthetic experiments and semi-synthetic experiments demonstrate the effectiveness of the inference and disentanglement techniques and our method achieves more accurate treatment effect estimation. © 2024 Institute of Physics Publishing. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

How Does A Text Preprocessing Pipeline Affect Ontology Syntactic Matching?

arXiv

引用

arXiv 2024年

作者： Qiang, Zhangcheng Taylor, Kerry Wang, Weiqing School of Computing Australian National University Canberra Australia Department of Data Science and AI Monash University Melbourne Australia

The generic text preprocessing pipeline, comprising Tokenisation, Normalisation, Stop Words Removal, and Stemming/ Lemmatisation, has been implemented in many ontology matching (OM) systems. However, the lack of standardisation in text preprocessing creates diversity in mapping results. In this paper, we investigate the effect of the text preprocessing pipeline on OM tasks at syntactic levels. Our experiments on 8 Ontology Alignment Evaluation Initiative (OAEI) track repositories with 49 distinct alignments indicate: (1) Tokenisation and Normalisation are currently more effective than Stop Words Removal and Stemming/Lemmatisation;and (2) The selection of Lemmatisation and Stemming is task-specific. We recommend standalone Lemmatisation or Stemming with post-hoc corrections. We find that (3) Porter Stemmer and Snowball Stemmer perform better than Lancaster Stemmer;and that (4) Part-of-Speech (POS) Tagging does not help Lemmatisation. To repair less effective Stop Words Removal and Stemming/Lemmatisation used in OM tasks, we propose a novel context-based pipeline repair approach that significantly improves matching correctness and overall matching performance. We also discuss the use of text preprocessing pipeline in the new era of large language models (LLMs). Copyright © 2024, The Authors. All rights reserved.

关键词： Syntactics

来源：评论

学校读者我要写书评

暂无评论

RVISA: Reasoning and Verification for Implicit Sentiment Analysis

引用

IEEE Transactions on Affective computing 2025年

作者： Lai, Wenna Xie, Haoran Xu, Guandong Li, Qing Hong Kong Polytechnic University Department of Computing Hong Kong Lingnan University School of Data Science Hong Kong University of Technology Sydney School of Computer Science Data Science Institute SydneyNSW2007 Australia Education University of Hong Kong Hong Kong

Under the context of the increasing social demand for fine-grained sentiment analysis (SA), implicit sentiment analysis (ISA) poses a significant challenge owing to the absence of salient cue words in expressions. Thus, reliable reasoning is required to understand how sentiment is evoked, enabling the identification of implicit sentiments. In the era of large language models (LLMs), encoder-decoder (ED) LLMs have emerged as popular backbone models for SA applications, given their impressive text comprehension and reasoning capabilities across diverse tasks. In comparison, decoder-only (DO) LLMs exhibit superior natural language generation and in-context learning capabilities. However, their responses may contain misleading or inaccurate information. To accurately identify implicit sentiments with reliable reasoning, this study introduces a two-stage reasoning framework named Reasoning and Verification for Implicit Sentiment Analysis (RVISA), which leverages the generation ability of DO LLMs and reasoning ability of ED LLMs to train an enhanced reasoner. The framework involves three-hop reasoning prompting to explicitly furnish sentiment elements as cues. The generated rationales are then used to fine-tune an ED LLM into a skilled reasoner. Additionally, we develop a straightforward yet effective answer-based verification mechanism to ensure the reliability of reasoning learning. Evaluation of the proposed method on two benchmark datasets demonstrates that it achieves state-of-the-art performance in ISA. © 2010-2012 IEEE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Large Language Models as Evolutionary Optimizers 13

Large Language Models as Evolutionary Optimizers

引用

13th IEEE Congress on Evolutionary Computation, CEC 2024

作者： Liu, Shengcai Chen, Caishun Qu, Xinghua Tang, Ke Ong, Yew-Soon Southern University of Science and Technology Centre for Frontier Ai Research A∗star Department of Computer Science and Engineering China Centre for Frontier Ai Research A∗star Singapore Tianqiao & Chrissy Chen Institute China Southern University of Science and Technology Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering China Nanyang Technological University Centre for Frontier Ai Research A∗star College of Computing and Data Science Singapore

ISBN: (纸本)9798350308365

Evolutionary algorithms (EAs) have achieved remarkable success in tackling complex combinatorial optimization problems. However, EAs often demand carefully-designed operators with the aid of domain expertise to achieve satisfactory performance. In this work, we present the first study on large language models (LLMs) as evolutionary combinatorial optimizers. The main advantage is that it requires minimal domain knowledge and human efforts, as well as no additional training of the model. This approach is referred to as LLM-driven EA (LMEA). Specifically, in each generation of the evolutionary search, LMEA instructs the LLM to select parent solutions from current population, and perform crossover and mutation to generate offspring solutions. Then, LMEA evaluates these new solutions and include them into the population for the next generation. LMEA is equipped with a self-adaptation mechanism that controls the temperature of the LLM. This enables it to balance between exploration and exploitation and prevents the search from getting stuck in local optima. We investigate the power of LMEA on the classical traveling salesman problems (TSPs) widely used in combinatorial optimization research. Notably, the results show that LMEA performs competitively to traditional heuristics in finding high-quality solutions on TSP instances with up to 20 nodes. Additionally, we also study the effectiveness of LLM-driven crossover/mutation and the self- adaptation mechanism in evolutionary search. In summary, our results reveal the great potentials of LLMs as evolutionary optimizers for solving combinatorial problems. We hope our research shall inspire future explorations on LLM-driven EAs for complex optimization challenges. © 2024 IEEE.

关键词： Traveling salesman problem

来源：评论

学校读者我要写书评

暂无评论

CareerBoost: A Hybrid RAG-NLP Job Recommendation Framework

CareerBoost: A Hybrid RAG-NLP Job Recommendation Framework

引用

International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)

作者： Sanskruti Deshmukh Aaryaman Bajaj Department of Artificial Intelligence and Data Science Vishwakarma Institute of Technology Pune India Department of Computing Technologies SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350376425

ISBN: (纸本)9798350376432

The proposed system presents a novel approach to job recommendation and resume enhancement, leveraging advanced Natural Language Processing (NLP) techniques and Retrieval-Augmented Generation (RAG). We introduce a comprehensive system that combines semantic search, FAISS-based similarity matching, and a fine-tuned GPT-2 model to provide personalized job recommendations and actionable resume improvement suggestions. The evaluation, conducted through offline metrics and online A/B testing with 5,000 users, demonstrates significant improvements over traditional keyword-matching approaches. The RAG-based system achieved a 25.8% increase in recommendation relevance (Precision@10), a 45.7% increase in user engagement, and a 41.2% increase in job application rates. Furthermore, users who applied the system’s resume enhancement suggestions saw a 9.0% improvement in subsequent job matches. These results highlight the potential of AI-driven systems to revolutionize online recruitment and career development, while also emphasizing the need for ongoing consideration of ethical implications and fairness in AI-assisted hiring processes.

关键词： Measurement Ethics Career development Accuracy Engineering profession Semantic search Resumes Natural language processing Recruitment Testing

来源：评论

学校读者我要写书评

暂无评论

ICPR 2024 Competition on Beyond Visible Spectrum: AI for Agriculture 27th

ICPR 2024 Competition on Beyond Visible Spectrum: AI for A...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Han, Liangxiu Huang, Wenjiang Zhang, Xin Dong, Yingying Sobeih, Tam Lin, Yufan Department of Computing and Mathematics Manchester Metropolitan University ManchesterM15 6BH United Kingdom Key laboratory of Digital Earth Science Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China Institute of Data Science National Cheng Kung University Taiwan70101 China

ISBN: (纸本)9783031801389

The ICPR 2024 Competition on "Beyond Visible Spectrum: AI for Agriculture" presents an exciting opportunity for researchers and practitioners to advance computer vision techniques in agricultural crop disease monitoring. By leveraging the vast multi/hyper spectral remote sensing datasets, participants are encouraged to develop innovative machines/deep learning algorithms. The primary objective is to improve the accuracy and efficiency of crop disease identification, thereby fostering sustainable agricultural practices and facilitating global knowledge generation for enhanced food security. This challenge presents participants with the opportunity to make significant contributions to the advancement of precision farming and crop monitoring techniques. By improving the machines/deep learning models, the competition aims to drive forward the capabilities of agricultural technology, making it more efficient, accurate, and sustainable. A total of 190 people from more than 10 countries in Europe, Asia and North America entered the competition by the deadline date, with a total of 126 submissions of results. The rank1 team obtained an accuracy of 73.3% in crop disease classification task with hyperspectral dataset. In the self-supervised learning task, the rank1 team obtained 79.2% accuracy. We report the leader-boards of the competition in this report and present three innovative approaches of the participants. The ICPR 2024 Competition on "Beyond Visible Spectrum" is not just a competition;it is a call to action for the computer research community to bring about transformative changes in how agricultural data is analyze and utilized, paving the way for smarter, more sustainable farming practices that can benefit communities worldwide. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：