← All research areas

LLMs, Multimodal & Agentic AI

Studying systems that interpret language, images and sensor signals, and the emerging behaviors of interacting AI agents.

Conceptual illustration of llms, multimodal & agentic ai

Research Description

Overview

We study LLMs, VLMs and multimodal models that interpret language, images and sensor information, alongside emerging agentic AI workflows. Key questions concern whether outputs remain grounded in input evidence, how modalities interact under attack and how privacy changes when people use AI assistants.

Published VLM robustness and consumer-assistant studies sit alongside emerging work on agent-to-agent interactions and multimodal screening prototypes. The goal is to characterize capabilities and failure modes with clear evaluation limits, rather than equating fluent responses with reliable decisions.

Research Thrusts

Grounded multimodal interpretation

Study how language, images and sensing signals contribute to model outputs, distinguishing evidence-grounded interpretation from fluent but unsupported responses.

Privacy and adversarial reliability

Evaluate LLM/VLM behavior in consumer tools and adversarial settings, including driving-scene understanding and writing-assistant privacy.

Emerging agent interactions

Explore agent-to-agent behavior and multimodal agent prototypes, with explicit attention to evaluation boundaries and edge–cloud constraints.

Research Goals

Evaluate multimodal capabilities and vulnerabilities, privacy in AI assistance, and agent interactions.

Intellectual Merit

Connect evidence across modalities and interaction settings, testing reliability instead of assuming that fluent output is correct.

Research Impact

The goal is a stronger basis for evaluating AI assistants and multimodal workflows before they are trusted in consequential settings. Exploratory agent and screening prototypes are not evidence of validated autonomous or clinical performance.

Current Projects & Emerging Directions

Existing research

Privacy in human–AI interaction

Text / audioConsumer AI transformationPrivacy ↔ utility

Everyday writing and audio-editing tools may reveal sensitive attributes, but may also help transform content to reduce unwanted inference. The same consumer interface can therefore act as both a privacy risk and a possible defense.

Research objectives

Measure how transformations affect sensitive-attribute inference while preserving the usefulness and meaning of the original content.

Approach & Intellectual Merit

Published writing-assistant and audio-editing studies motivate controlled comparisons across original and transformed inputs. Privacy protection, task utility and model-specific behavior must be assessed together; fluent rewrites alone do not demonstrate protection.

Selected publications

No Innocence in Styling: Discovery of Privacy Protection Capabilities and Security Risks in Consumer Generative AI Writing Assistants ↗

Relevance & Applications

Relevant settings include consumer platforms, communication tools and device-integrated assistants using LLMs. Results support informed evaluation of privacy tradeoffs, not universal guarantees across people, models or platforms.

Impact

The published work examines both privacy-protection capabilities and security risks. It may help researchers and users assess tradeoffs, but does not establish universal privacy guarantees or measured adoption outcomes.

Existing research

Language models for cyber-physical systems

Physical observationsLanguage-assisted interpretationHuman oversight

Language models can connect sensor observations with human-readable descriptions and higher-level decisions. Physical environments also impose constraints that text-only benchmarks may miss, including timing, incomplete measurements and consequences of erroneous outputs.

Research objectives

Explore useful language-assisted sensing and interpretation while preserving privacy, reliability and resource awareness.

Approach & Intellectual Merit

Published architectures and fall-detection studies provide foundations, alongside wearable-data integrity research. The evaluation agenda connects sensor evidence to model output and considers uncertainty and human oversight rather than assuming autonomous control is safe.

Selected publications

Adaptive and Robust Data Poisoning Detection and Sanitization in Wearable IoT Systems using Large Language Models ↗

Relevance & Applications

Applications include smart environments, wearable assistance and healthcare-oriented research prototypes. Each setting requires separate testing of detection errors, response behavior and user impact.

Impact

The aim is to connect useful language-model capabilities with the privacy, security, and resource requirements of physical environments. The linked wearable-data study is supporting research, not evidence of clinical deployment or validated autonomous control.

Published foundations and emerging preprints

Multimodal robustness and agent interactions

Multimodal evidenceModels & agent interactionsRobustness assessment

Vision–language models combine complementary evidence but can be vulnerable when an input is perturbed or modalities disagree. Interacting agents add another layer: one system’s outputs can become another system’s context or decision input.

Research objectives

Characterize robustness across modalities and investigate behavior across agent-to-agent interactions.

Approach & Intellectual Merit

Published VLM robustness studies and emerging agent-interaction preprints form distinct evidence streams. The evaluation agenda connects input evidence, model response and interaction outcomes while keeping prototype results separate from deployment claims.

Relevance & Applications

Relevant research settings include driving scenes, consumer AI and the VetClaw veterinary screening prototype. These studies do not establish validated autonomous driving, clinical screening performance or general-purpose agent reliability.

Impact

Evidence for evaluating vulnerabilities and prototype behavior, not validated autonomous driving or clinical performance.

Related Publications

CVPR WorkshopWorkshop

Breaking Sensor-Text Alignment: Cross-Modal Attack on Contrastive Multimodal Human Activity Recognition

Malithi Mithsara Wanniarachchi Kankanamge, Ning Yang, Ahmed Imteaj, Abdur R. Shahid

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops · pp. 3966-3975, 2026

AI Security & PrivacyMultimodal AIWearable Sensing
PDFPaper ↗
BibTeX
@inproceedings{shield73,
  title = {Breaking Sensor-Text Alignment: Cross-Modal Attack on Contrastive Multimodal Human Activity Recognition},
  author = {Malithi Mithsara Wanniarachchi Kankanamge and Ning Yang and Ahmed Imteaj and Abdur R. Shahid},
  year = {2026},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
  pages = {3966-3975},
  url = {https://openaccess.thecvf.com/content/CVPR2026W/SenseOfSpace/html/Kankanamge_Breaking_Sensor-Text_Alignment_Cross-Modal_Attack_on_Contrastive_Multimodal_Human_Activity_CVPRW_2026_paper.html}
}
ICORE A*2026 editionConference

Edits in the Wild: A Multilingual Benchmark for Localizing On-Device Generative Image Edits on Consumer Platforms

Mohd. Farhan Israk Soumik, Awal Ahmed Fime, Malithi Mithsara Wanniarachchi Kankanamge, Arun Babu Aryal, Hoda Joudi Haghighi, Bibek Koirala, Ahmed Imteaj, Abdur R. Shahid

Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing · pp. 37756–37789, 2026

Digital ForensicsComputer VisionMultimodal AI
Paper ↗
BibTeX
@inproceedings{shield86,
  title = {Edits in the Wild: A Multilingual Benchmark for Localizing On-Device Generative Image Edits on Consumer Platforms},
  author = {Mohd. Farhan Israk Soumik and Awal Ahmed Fime and Malithi Mithsara Wanniarachchi Kankanamge and Arun Babu Aryal and Hoda Joudi Haghighi and Bibek Koirala and Ahmed Imteaj and Abdur R. Shahid},
  year = {2026},
  booktitle = {Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing},
  pages = {37756–37789},
  url = {https://aclanthology.org/2026.emnlp-main.1661/}
}
Preprint

Evaluating Federated Vision-Language and Vision-Only Models for Privacy-Aware Breast Cancer Imaging

Malithi Mithsara Wanniarachchi Kankanamge, Ning Yang, Sree Bash Debnath, Md Zobaer Islam, Abdur R Shahid

arXiv preprint arXiv:2610.36379 · 2026

Federated LearningHealthcare AIMultimodal AI
PDFPaper ↗
BibTeX
@misc{shield88,
  title = {Evaluating Federated Vision-Language and Vision-Only Models for Privacy-Aware Breast Cancer Imaging},
  author = {Malithi Mithsara Wanniarachchi Kankanamge and Ning Yang and Sree Bash Debnath and Md Zobaer Islam and Abdur R Shahid},
  year = {2026},
  howpublished = {arXiv preprint arXiv:2610.36379},
  url = {https://arxiv.org/abs/2610.36379}
}
Preprint

From Notepad AI to Social Media: How Can Text Style Transformation Mitigate Social Harm?

Syed Mhamudul Hasan, Mohd Soumik, Abdur R Shahid

arXiv preprint arXiv:2604.27365 · 2026

LLMs
PDFPaper ↗
BibTeX
@misc{shield77,
  title = {From Notepad AI to Social Media: How Can Text Style Transformation Mitigate Social Harm?},
  author = {Syed Mhamudul Hasan and Mohd Soumik and Abdur R Shahid},
  year = {2026},
  howpublished = {arXiv preprint arXiv:2604.27365},
  url = {https://arxiv.org/abs/2604.27365}
}
Preprint

Modeling Emotional Dynamics in Agent-to-Agent Interactions on Moltbook

Syed Mhamudul Hasan, Abdur R Shahid

arXiv preprint arXiv:2605.20442 · 2026

Agentic AI
PDFPaper ↗
BibTeX
@misc{shield78,
  title = {Modeling Emotional Dynamics in Agent-to-Agent Interactions on Moltbook},
  author = {Syed Mhamudul Hasan and Abdur R Shahid},
  year = {2026},
  howpublished = {arXiv preprint arXiv:2605.20442},
  url = {https://arxiv.org/abs/2605.20442}
}
Conference

No Innocence in Styling: Discovery of Privacy Protection Capabilities and Security Risks in Consumer Generative AI Writing Assistants

Mohd. Farhan Israk Soumik, Syed Mhamudul Hasan, Wanniarachchi Kankanamge Malithi Mithsara, Ahmed Imteaj, Abdur R. Shahid

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track) · pp. 1866–1885, 2026

AI Security & PrivacyLLMs
PDFPaper ↗DOI ↗
BibTeX
@inproceedings{shield75,
  title = {No Innocence in Styling: Discovery of Privacy Protection Capabilities and Security Risks in Consumer Generative AI Writing Assistants},
  author = {Mohd. Farhan Israk Soumik and Syed Mhamudul Hasan and Wanniarachchi Kankanamge Malithi Mithsara and Ahmed Imteaj and Abdur R. Shahid},
  year = {2026},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)},
  pages = {1866–1885},
  publisher = {Association for Computational Linguistics},
  doi = {10.18653/v1/2026.acl-industry.128},
  url = {https://aclanthology.org/2026.acl-industry.128/}
}
JIF 4.92025 metricJournal

Secure Yet Fragile: Adversarial Vulnerabilities of Federated Vision–Language Models in Medical AI

A.A. Fime, T.Z. Samiha, M.Z. Hossain, S. Zaman, A.M. Shibli, A.R. Shahid, Zhen Ni, A. Imteaj

Nature Scientific Reports Journal · 2026

AI Security & PrivacyFederated LearningMultimodal AI
Paper ↗DOI ↗
BibTeX
@article{shield72,
  title = {Secure Yet Fragile: Adversarial Vulnerabilities of Federated Vision–Language Models in Medical AI},
  author = {A.A. Fime and T.Z. Samiha and M.Z. Hossain and S. Zaman and A.M. Shibli and A.R. Shahid and Zhen Ni and A. Imteaj},
  year = {2026},
  journal = {Nature Scientific Reports Journal},
  publisher = {Nature},
  doi = {10.1038/s41598-026-48102-4},
  url = {https://doi.org/10.1038/s41598-026-48102-4}
}
Preprint

VetClaw: An edge-cloud multimodal agentic system for veterinary disease screening

Syed Mhamudul Hasan, Anas AlSobeh, Hussein Zangoti, Abdur R Shahid

arXiv preprint arXiv:2607.26042 · 2026

Agentic AIMultimodal AIHealthcare AI
PDFPaper ↗
BibTeX
@misc{shield84,
  title = {VetClaw: An edge-cloud multimodal agentic system for veterinary disease screening},
  author = {Syed Mhamudul Hasan and Anas AlSobeh and Hussein Zangoti and Abdur R Shahid},
  year = {2026},
  howpublished = {arXiv preprint arXiv:2607.26042},
  url = {https://arxiv.org/abs/2607.26042}
}
Preprint

A comprehensive survey of large language models for human activity recognition: opportunities, challenges, and future directions

Malithi Mithsara Wanniarachchi Kankanamge, Syed Mhamudul Hasan, Hussein Zangoti, Ahmed Imteaj, Ning Yang, Abdur R Shahid

TechRxiv · 2025

LLMsWearable Sensing
PDFPaper ↗DOI ↗
BibTeX
@misc{shield68,
  title = {A comprehensive survey of large language models for human activity recognition: opportunities, challenges, and future directions},
  author = {Malithi Mithsara Wanniarachchi Kankanamge and Syed Mhamudul Hasan and Hussein Zangoti and Ahmed Imteaj and Ning Yang and Abdur R Shahid},
  year = {2025},
  howpublished = {TechRxiv},
  publisher = {TechRxiv},
  doi = {10.36227/techrxiv.176162216.64823146/v1},
  url = {https://doi.org/10.36227/techrxiv.176162216.64823146/v1}
}
Preprint

Adaptive and Robust Data Poisoning Detection and Sanitization in Wearable IoT Systems using Large Language Models

W. K. M Mithsara, Ning Yang, Ahmed Imteaj, Hussein Zangoti, Abdur R. Shahid

arXiv · 2025

AI Security & PrivacyLLMsWearable Sensing
PDFPaper ↗
BibTeX
@misc{shield69,
  title = {Adaptive and Robust Data Poisoning Detection and Sanitization in Wearable IoT Systems using Large Language Models},
  author = {W. K. M Mithsara and Ning Yang and Ahmed Imteaj and Hussein Zangoti and Abdur R. Shahid},
  year = {2025},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2511.02894}
}
Workshop

Audio Editing Features As User-Centric Privacy Defenses Against Large Language Model (LLM)-Based Emotion Inference Attacks

Mohd. Farhan Israk Soumik, W.K.M Mithsara, Abdur Rahman Bin Shahid, Ahmed Imteaj

The Sixth AAAI-25 Workshop on Privacy-Preserving Artificial Intelligence · 2025

AI Security & PrivacyLLMs
PDFPaper ↗
BibTeX
@inproceedings{shield55,
  title = {Audio Editing Features As User-Centric Privacy Defenses Against Large Language Model (LLM)-Based Emotion Inference Attacks},
  author = {Mohd. Farhan Israk Soumik and W.K.M Mithsara and Abdur Rahman Bin Shahid and Ahmed Imteaj},
  year = {2025},
  booktitle = {The Sixth AAAI-25 Workshop on Privacy-Preserving Artificial Intelligence},
  publisher = {https://arxiv.org/abs/2501.18727},
  url = {https://arxiv.org/abs/2501.18727}
}
Symposium

Benchmarking large-language models for resource-efficient medical AI for edge deployment

Awal Ahmed Fime, Md Zarif Hossain, Saika Zaman, Abdur Rahman Bin Shahid, Ahmed Imteaj

Proceedings of the AAAI Symposium Series · vol. 5, no. 1, pp. 137-142, 2025

LLMsSustainable & Resource-Efficient AIHealthcare AI
Summary
This paper compares parameter-efficient fine-tuning of large language models for medical question answering, considering answer quality and resource efficiency for edge deployment.
PDFPaper ↗DOI ↗
BibTeX
@inproceedings{shield59,
  title = {Benchmarking large-language models for resource-efficient medical AI for edge deployment},
  author = {Awal Ahmed Fime and Md Zarif Hossain and Saika Zaman and Abdur Rahman Bin Shahid and Ahmed Imteaj},
  year = {2025},
  booktitle = {Proceedings of the AAAI Symposium Series},
  volume = {5},
  number = {1},
  pages = {137-142},
  doi = {10.1609/aaaiss.v5i1.35580},
  url = {https://doi.org/10.1609/aaaiss.v5i1.35580}
}
Conference

Breaking and Securing Vision-Language Models: An Adversarial Robustness Study

Md Zarif Hossain, Abdur R Shahid, Ahmed Imteaj

2025 International Conference on Machine Learning and Applications (ICMLA) · pp. 1411-1416, 2025

AI Security & PrivacyMultimodal AI
Paper ↗DOI ↗
BibTeX
@inproceedings{shield71,
  title = {Breaking and Securing Vision-Language Models: An Adversarial Robustness Study},
  author = {Md Zarif Hossain and Abdur R Shahid and Ahmed Imteaj},
  year = {2025},
  booktitle = {2025 International Conference on Machine Learning and Applications (ICMLA)},
  pages = {1411-1416},
  publisher = {IEEE},
  doi = {10.1109/ICMLA66185.2025.00215},
  url = {https://doi.org/10.1109/ICMLA66185.2025.00215}
}
Conference

Development of a Fall Detection and Safety Communication System Using Small Language Models

Malithi Wanniarachchi Kankanamge, Abdur R Shahid, Ning Yang, SM Jamil Uddin, Rahul Biswas

ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction · vol. 42, pp. 588-594, 2025

LLMsCyber-Physical Systems & IoT
PDFPaper ↗DOI ↗
BibTeX
@inproceedings{shield65,
  title = {Development of a Fall Detection and Safety Communication System Using Small Language Models},
  author = {Malithi Wanniarachchi Kankanamge and Abdur R Shahid and Ning Yang and SM Jamil Uddin and Rahul Biswas},
  year = {2025},
  booktitle = {ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction},
  volume = {42},
  pages = {588-594},
  publisher = {IAARC Publications},
  doi = {10.22260/ISARC2025/0077},
  url = {https://doi.org/10.22260/ISARC2025/0077}
}
Preprint

Evaluating Apple Intelligence's Writing Tools for Privacy Against Large Language Model-Based Inference Attacks: Insights from Early Datasets

Mohd. Farhan Israk Soumik, Syed Mhamudul Hasan, Abdur R. Shahid

arXiv preprint arXiv:2506.03870 · 2025

AI Security & PrivacyLLMs
PDFPaper ↗DOI ↗
BibTeX
@misc{shield62,
  title = {Evaluating Apple Intelligence's Writing Tools for Privacy Against Large Language Model-Based Inference Attacks: Insights from Early Datasets},
  author = {Mohd. Farhan Israk Soumik and Syed Mhamudul Hasan and Abdur R. Shahid},
  year = {2025},
  howpublished = {arXiv preprint arXiv:2506.03870},
  doi = {10.48550/arXiv.2506.03870},
  url = {https://arxiv.org/abs/2506.03870}
}
Conference

How Reliable Are Large Language Models? Zero-Shot Detection of Construction Hazards

Nishi Chaudhary, SM Jamil Uddin, Mahzabin Tamanna, Alex Albert, Abdur Rahman Bin Shahid

ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction · vol. 42, pp. 649-656, 2025

LLMs
Paper ↗DOI ↗
BibTeX
@inproceedings{shield67,
  title = {How Reliable Are Large Language Models? Zero-Shot Detection of Construction Hazards},
  author = {Nishi Chaudhary and SM Jamil Uddin and Mahzabin Tamanna and Alex Albert and Abdur Rahman Bin Shahid},
  year = {2025},
  booktitle = {ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction},
  volume = {42},
  pages = {649-656},
  publisher = {IAARC Publications},
  doi = {10.22260/ISARC2025/0085},
  url = {https://doi.org/10.22260/ISARC2025/0085}
}
Conference

Privacy-Preserving Multimodal Stress Detection from Wearables with Attention Fusion and Federated Learning

Mohd Farhan Israk Soumik, Hussein Zangoti, Ebrahim Maghsoudlou, Awal Ahmed Fime, Ahmed Imteaj, Abdur R Shahid

2025 International Conference on Machine Learning and Applications (ICMLA) · pp. 1369-1375, 2025

AI Security & PrivacyFederated LearningWearable Sensing
Paper ↗DOI ↗
BibTeX
@inproceedings{shield74,
  title = {Privacy-Preserving Multimodal Stress Detection from Wearables with Attention Fusion and Federated Learning},
  author = {Mohd Farhan Israk Soumik and Hussein Zangoti and Ebrahim Maghsoudlou and Awal Ahmed Fime and Ahmed Imteaj and Abdur R Shahid},
  year = {2025},
  booktitle = {2025 International Conference on Machine Learning and Applications (ICMLA)},
  pages = {1369-1375},
  publisher = {IEEE},
  doi = {10.1109/ICMLA66185.2025.00209},
  url = {https://doi.org/10.1109/ICMLA66185.2025.00209}
}
CVPR WorkshopWorkshop

Towards trustworthy autonomous vehicles with vision-language models under targeted and untargeted adversarial attacks

Awal Ahmed Fime, Md Zarif Hossain, Saika Zaman, Abdur R Shahid, Ahmed Imteaj

2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · pp. 619-628, 2025

AI Security & PrivacyMultimodal AIComputer Vision
Paper ↗
BibTeX
@inproceedings{shield63,
  title = {Towards trustworthy autonomous vehicles with vision-language models under targeted and untargeted adversarial attacks},
  author = {Awal Ahmed Fime and Md Zarif Hossain and Saika Zaman and Abdur R Shahid and Ahmed Imteaj},
  year = {2025},
  booktitle = {2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  pages = {619-628},
  publisher = {IEEE},
  url = {https://openaccess.thecvf.com/content/CVPR2025W/TCV/html/Fime_Towards_Trustworthy_Autonomous_Vehicles_with_Vision-Language_Models_Under_Targeted_and_CVPRW_2025_paper.html}
}
Conference Poster

Detection of Poisoning Attacks in Human Activity Recognition using Large Language Models

W.K.M Mithsara, Abdur R. Shahid, Ning Yang

IEEE-EMBS International Conference on Body Sensor Networks (IEEE BSN) · 2024

AI Security & PrivacyLLMsWearable Sensing
Conference program ↗
BibTeX
@misc{shield49,
  title = {Detection of Poisoning Attacks in Human Activity Recognition using Large Language Models},
  author = {W.K.M Mithsara and Abdur R. Shahid and Ning Yang},
  year = {2024},
  howpublished = {IEEE-EMBS International Conference on Body Sensor Networks (IEEE BSN)}
}
Conference

Distributed threat intelligence at the edge devices: A large language model-driven approach

Syed Mhamudul Hasan, Alaa M Alotaibi, Sajedul Talukder, Abdur R Shahid

2024 IEEE 48th Annual Computers, Software, and Applications Conference (COMPSAC) · pp. 1496-1497, 2024

AI Security & PrivacyLLMs
Paper ↗DOI ↗
BibTeX
@inproceedings{shield40,
  title = {Distributed threat intelligence at the edge devices: A large language model-driven approach},
  author = {Syed Mhamudul Hasan and Alaa M Alotaibi and Sajedul Talukder and Abdur R Shahid},
  year = {2024},
  booktitle = {2024 IEEE 48th Annual Computers, Software, and Applications Conference (COMPSAC)},
  pages = {1496-1497},
  publisher = {IEEE},
  doi = {10.1109/compsac61105.2024.00206},
  url = {https://doi.org/10.1109/compsac61105.2024.00206}
}
Preprint

From Early Adoption to Ethical Adoption: A Diffusion of Innovation Perspective on ChatGPT and Large Language Models in the Classroom

Mohammad Imran, Abdur R. Shahid, Minghui Hou, Ahmed Imteaj

TechRxiv · 2024

EducationLLMs
PDFDOI ↗
BibTeX
@misc{shield34,
  title = {From Early Adoption to Ethical Adoption: A Diffusion of Innovation Perspective on ChatGPT and Large Language Models in the Classroom},
  author = {Mohammad Imran and Abdur R. Shahid and Minghui Hou and Ahmed Imteaj},
  year = {2024},
  howpublished = {TechRxiv},
  doi = {10.36227/techrxiv.170630660.06963201/v1}
}
Conference

Intelligent Fall Detection and Emergency Response for Smart Homes Using Language Models

W.K.M Mithsara, Abdur R. Shahid, Ning Yang

23rd International Conference on Machine Learning and Applications (ICMLA 2024) · 2024

LLMsCyber-Physical Systems & IoT
Paper ↗DOI ↗
BibTeX
@inproceedings{shield52,
  title = {Intelligent Fall Detection and Emergency Response for Smart Homes Using Language Models},
  author = {W.K.M Mithsara and Abdur R. Shahid and Ning Yang},
  year = {2024},
  booktitle = {23rd International Conference on Machine Learning and Applications (ICMLA 2024)},
  doi = {10.1109/ICMLA61862.2024.00037},
  url = {https://doi.org/10.1109/ICMLA61862.2024.00037}
}
Conference

Large Language Model Integrated Healthcare Cyber-Physical Systems Architecture

Malithi Wanniarachchi Kankanamge, Syed Mhamudul Hasan, Abdur R. Shahid, Ning Yang

48th IEEE International Conference on Computers, Software, and Applications (COMPSAC 2024) · 2024

LLMsCyber-Physical Systems & IoTHealthcare AI
Paper ↗DOI ↗
BibTeX
@inproceedings{shield45,
  title = {Large Language Model Integrated Healthcare Cyber-Physical Systems Architecture},
  author = {Malithi Wanniarachchi Kankanamge and Syed Mhamudul Hasan and Abdur R. Shahid and Ning Yang},
  year = {2024},
  booktitle = {48th IEEE International Conference on Computers, Software, and Applications (COMPSAC 2024)},
  doi = {10.1109/compsac61105.2024.00228},
  url = {https://doi.org/10.1109/compsac61105.2024.00228}
}
NeurIPS Workshop · PosterPoster / abstract

Poster: Leveraging Large Language Models for Zero-Shot Detection and Mitigation of Data Poisoning in Wearable AI Systems

Malithi Wanniarachchi Kankanamge, Abdur R Shahid, Ning Yang

NeurIPS workshop on GenAI for Health: Potential, Trust and Policy Compliance · 2024

AI Security & PrivacyLLMsWearable Sensing
PDFPaper ↗
BibTeX
@misc{shield54,
  title = {Poster: Leveraging Large Language Models for Zero-Shot Detection and Mitigation of Data Poisoning in Wearable AI Systems},
  author = {Malithi Wanniarachchi Kankanamge and Abdur R Shahid and Ning Yang},
  year = {2024},
  howpublished = {NeurIPS workshop on GenAI for Health: Potential, Trust and Policy Compliance},
  url = {https://openreview.net/forum?id=GAEZOAqrtA}
}
JIF 6.12025 metricJournal

Sticks and Stones May Break My Bones, But Words Will Never Hurt Me! — Navigating the Cybersecurity Risks of Generative AI

Abdur R. Shahid, Ahmed Imteaj

AI & Society · 2024

AI Security & PrivacyLLMs
Paper ↗DOI ↗
BibTeX
@article{shield38,
  title = {Sticks and Stones May Break My Bones, But Words Will Never Hurt Me! — Navigating the Cybersecurity Risks of Generative AI},
  author = {Abdur R. Shahid and Ahmed Imteaj},
  year = {2024},
  journal = {AI & Society},
  publisher = {Springer},
  doi = {10.1007/s00146-024-01934-y},
  url = {https://doi.org/10.1007/s00146-024-01934-y}
}
Conference

Tripleplay: Enhancing federated learning with clip for non-iid data and resource efficiency

Ahmed Imteaj, Md Zarif Hossain, Saika Zaman, Abdur R Shahid

2024 International Conference on Machine Learning and Applications (ICMLA) · pp. 1474-1480, 2024

Federated LearningMultimodal AISustainable & Resource-Efficient AI
Paper ↗DOI ↗
BibTeX
@inproceedings{shield50,
  title = {Tripleplay: Enhancing federated learning with clip for non-iid data and resource efficiency},
  author = {Ahmed Imteaj and Md Zarif Hossain and Saika Zaman and Abdur R Shahid},
  year = {2024},
  booktitle = {2024 International Conference on Machine Learning and Applications (ICMLA)},
  pages = {1474-1480},
  publisher = {IEEE},
  doi = {10.1109/ICMLA61862.2024.00228},
  url = {https://doi.org/10.1109/ICMLA61862.2024.00228}
}
Conference

WatchOverGPT: A Framework for Real-Time Crime Detection and Response Using Wearable Camera and Large Language Model (LLM)

Abdur R. Shahid, Syed Mhamudul Hasan, Malithi Wanniarachchi Kankanamge, Md Zarif Hossain, Ahmed Imteaj

48th IEEE International Conference on Computers, Software, and Applications (COMPSAC 2024) · 2024

LLMsWearable SensingComputer Vision
Paper ↗DOI ↗
BibTeX
@inproceedings{shield46,
  title = {WatchOverGPT: A Framework for Real-Time Crime Detection and Response Using Wearable Camera and Large Language Model (LLM)},
  author = {Abdur R. Shahid and Syed Mhamudul Hasan and Malithi Wanniarachchi Kankanamge and Md Zarif Hossain and Ahmed Imteaj},
  year = {2024},
  booktitle = {48th IEEE International Conference on Computers, Software, and Applications (COMPSAC 2024)},
  doi = {10.1109/compsac61105.2024.00351},
  url = {https://doi.org/10.1109/COMPSAC61105.2024.00351}
}
Conference

Zero-Shot Detection and Sanitization of Data Poisoning Attacks in Wearable AI using Large Language Models

W.K.M Mithsara, Abdur R. Shahid, Ning Yang

2024 International Conference on Machine Learning and Applications (ICMLA) · 2024

AI Security & PrivacyLLMsWearable Sensing
Paper ↗DOI ↗
BibTeX
@inproceedings{shield53,
  title = {Zero-Shot Detection and Sanitization of Data Poisoning Attacks in Wearable AI using Large Language Models},
  author = {W.K.M Mithsara and Abdur R. Shahid and Ning Yang},
  year = {2024},
  booktitle = {2024 International Conference on Machine Learning and Applications (ICMLA)},
  doi = {10.1109/ICMLA61862.2024.00233},
  url = {https://doi.org/10.1109/ICMLA61862.2024.00233}
}