Pub Date : 2025-01-01Epub Date: 2025-05-08DOI: 10.1007/s44196-025-00853-0
Mohammad Reza Zarei, Abbas Akkasi, Majid Komeili
Vision-language models (VLMs) pre-trained on large-scale image-text pairs have shown impressive results in zero-shot vision tasks. Knowledge transferability of these models can be further improved with the help of a limited number of samples. Feature adapter tuning is a prominent approach employed for efficient transfer learning (ETL). However, most of the previous ETL models focus on tuning either prior-independent or prior-dependent feature adapters. We propose a novel ETL approach that leverages both adapter styles simultaneously. Additionally, most existing ETL models rely on using textual prompts constructed by completing general pre-defined templates. This approach neglects the descriptive knowledge that can assist VLM by presenting an informative prompt. Instead of pre-defined templates for prompt construction, we use a pre-trained LLM to generate attribute-specific prompts for each visual category. Furthermore, we guide the VLM with context-aware discriminative information generated by the pre-trained LLM to emphasize features that distinguish the most probable candidate classes. The proposed ETL model is evaluated on 11 datasets and sets a new state of the art. Our code and all collected prompts are publicly available at https://github.com/mrzarei5/DATViL.
{"title":"Dual Adapter Tuning of Vision-Language Models Using Large Language Models.","authors":"Mohammad Reza Zarei, Abbas Akkasi, Majid Komeili","doi":"10.1007/s44196-025-00853-0","DOIUrl":"https://doi.org/10.1007/s44196-025-00853-0","url":null,"abstract":"<p><p>Vision-language models (VLMs) pre-trained on large-scale image-text pairs have shown impressive results in zero-shot vision tasks. Knowledge transferability of these models can be further improved with the help of a limited number of samples. Feature adapter tuning is a prominent approach employed for efficient transfer learning (ETL). However, most of the previous ETL models focus on tuning either prior-independent or prior-dependent feature adapters. We propose a novel ETL approach that leverages both adapter styles simultaneously. Additionally, most existing ETL models rely on using textual prompts constructed by completing general pre-defined templates. This approach neglects the descriptive knowledge that can assist VLM by presenting an informative prompt. Instead of pre-defined templates for prompt construction, we use a pre-trained LLM to generate attribute-specific prompts for each visual category. Furthermore, we guide the VLM with context-aware discriminative information generated by the pre-trained LLM to emphasize features that distinguish the most probable candidate classes. The proposed ETL model is evaluated on 11 datasets and sets a new state of the art. Our code and all collected prompts are publicly available at https://github.com/mrzarei5/DATViL.</p>","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"18 1","pages":"109"},"PeriodicalIF":2.9,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12077310/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144081913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-01Epub Date: 2025-11-03DOI: 10.1007/s44196-025-01025-w
Raed Alharthi, Stephen Ojo, Thomas I Nathaniel, Nagwan Abdel Samee, Muhammad Umer, Mona M Jamjoom, Shtwai Alsubai, Jawad Khan
The proliferation of AI-generated text, fueled by large language models (LLMs), presents pressing challenges in maintaining content authenticity, safeguarding academic integrity, and mitigating misinformation. This paper proposes a responsible detection and mitigation framework that leverages hybrid neural networks and multi-feature fusion to distinguish AI-generated text from human-authored content. The proposed model integrates BERT-based semantic embeddings, convolutional features via Text-CNN, and statistical descriptors into a unified representation. A CNN-BiLSTM architecture is employed to capture both local syntactic patterns and long-range semantic dependencies. The framework emphasizes responsible AI (RAI) by prioritizing interpretability and reducing bias in detection decisions. Extensive evaluations on a balanced benchmark dataset demonstrate the model's superior performance, achieving 95.4% accuracy, 94.8% precision, 94.1% recall, and a 96.7% F1-score-outperforming leading transformer-based baselines. The proposed framework is also evaluated on the CoAID external independent dataset to show generalizability. This study contributes to the responsible deployment of LLMs by enhancing transparency and robustness in AI-generated content verification, paving the way for secure and ethical integration of generative models into content management systems.
{"title":"Responsible Detection and Mitigation of AI-Generated Text Using Hybrid Neural Networks and Feature Fusion: Toward Trustworthy Content Management in the Era of Large Language Models.","authors":"Raed Alharthi, Stephen Ojo, Thomas I Nathaniel, Nagwan Abdel Samee, Muhammad Umer, Mona M Jamjoom, Shtwai Alsubai, Jawad Khan","doi":"10.1007/s44196-025-01025-w","DOIUrl":"10.1007/s44196-025-01025-w","url":null,"abstract":"<p><p>The proliferation of AI-generated text, fueled by large language models (LLMs), presents pressing challenges in maintaining content authenticity, safeguarding academic integrity, and mitigating misinformation. This paper proposes a responsible detection and mitigation framework that leverages hybrid neural networks and multi-feature fusion to distinguish AI-generated text from human-authored content. The proposed model integrates BERT-based semantic embeddings, convolutional features via Text-CNN, and statistical descriptors into a unified representation. A CNN-BiLSTM architecture is employed to capture both local syntactic patterns and long-range semantic dependencies. The framework emphasizes responsible AI (RAI) by prioritizing interpretability and reducing bias in detection decisions. Extensive evaluations on a balanced benchmark dataset demonstrate the model's superior performance, achieving 95.4% accuracy, 94.8% precision, 94.1% recall, and a 96.7% F1-score-outperforming leading transformer-based baselines. The proposed framework is also evaluated on the CoAID external independent dataset to show generalizability. This study contributes to the responsible deployment of LLMs by enhancing transparency and robustness in AI-generated content verification, paving the way for secure and ethical integration of generative models into content management systems.</p>","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"18 1","pages":"274"},"PeriodicalIF":3.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12583390/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145454116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A LDA-Based Social Media Data Mining Framework for Plastic Circular Economy","authors":"Yangyimin Xue, Chandrasekhar Kambhampati, Yongqiang Cheng, Nishikant Mishra, N. Wulandhari, Pauline Deutz","doi":"10.1007/s44196-023-00375-7","DOIUrl":"https://doi.org/10.1007/s44196-023-00375-7","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"89 9","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139440433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-10DOI: 10.1007/s44196-023-00383-7
L. Nosrati, A. Bidgoli, H. Javadi
{"title":"Identifying People’s Faces in Smart Banking Systems Using Artificial Neural Networks","authors":"L. Nosrati, A. Bidgoli, H. Javadi","doi":"10.1007/s44196-023-00383-7","DOIUrl":"https://doi.org/10.1007/s44196-023-00383-7","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"36 10","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139441397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-08DOI: 10.1007/s44196-023-00391-7
M. A. Salman, M. A. Mahdi, Samaher Al-Janabi
{"title":"A GMEE-WFED System: Optimizing Wind Turbine Distribution for Enhanced Renewable Energy Generation in the Future","authors":"M. A. Salman, M. A. Mahdi, Samaher Al-Janabi","doi":"10.1007/s44196-023-00391-7","DOIUrl":"https://doi.org/10.1007/s44196-023-00391-7","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"20 17","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139445652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-08DOI: 10.1007/s44196-023-00389-1
Dongfang Zhao, Huanshi Xu, Zhang Xun
{"title":"Active Exploration Deep Reinforcement Learning for Continuous Action Space with Forward Prediction","authors":"Dongfang Zhao, Huanshi Xu, Zhang Xun","doi":"10.1007/s44196-023-00389-1","DOIUrl":"https://doi.org/10.1007/s44196-023-00389-1","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"56 18","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139447358","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-03DOI: 10.1007/s44196-023-00379-3
Li Song
{"title":"Design and Implementation of Remote Piano Teaching Based on Attention-Induced Multi-Head Convolutional Neural Network Optimized with Hunter–Prey Optimization","authors":"Li Song","doi":"10.1007/s44196-023-00379-3","DOIUrl":"https://doi.org/10.1007/s44196-023-00379-3","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"142 25","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139387450","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-03DOI: 10.1007/s44196-023-00388-2
Punit Gupta, F. Rustam, Khadija Kanwal, Wajdi Aljedaani, Sultan Alfarhood, Mejdl S. Safran, I. Ashraf
{"title":"Detecting Thyroid Disease Using Optimized Machine Learning Model Based on Differential Evolution","authors":"Punit Gupta, F. Rustam, Khadija Kanwal, Wajdi Aljedaani, Sultan Alfarhood, Mejdl S. Safran, I. Ashraf","doi":"10.1007/s44196-023-00388-2","DOIUrl":"https://doi.org/10.1007/s44196-023-00388-2","url":null,"abstract":"","PeriodicalId":54967,"journal":{"name":"International Journal of Computational Intelligence Systems","volume":"142 48","pages":""},"PeriodicalIF":2.9,"publicationDate":"2024-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139387422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号