Repository for the journal article, 'Federated Semi-Supervised Multi-Task Learning to Detect COVID-19 and Lungs Segmentation Marking Using Chest Radiography Images and Raspberry Pi Devices: An Internet of Medical Things Application', Mahbub Ul Alam, Rahim Rahmani. Sensors 21, no. 15: 5025, https://doi.org/10.3390/s21155025.

Visual Entities Empowered Zero-Shot Image-to-Text Generation Transfer Across Domains

Implementation of GATS from the paper: "GATS: Gather-Attend-Scatter" in pytorch and zeta

Implementation of Qformer from BLIP2 in Zeta Lego blocks.

Implementation of Attention on Attention in Zeta

This is the official repository for Vista dataset - A Vietnamese multimodal dataset contains more than 700,000 samples of conversations and images

Repository for the conference paper 'COVID-19 detection from thermal image and tabular medical data utilizing multi-modal machine learning', Mahbub Ul Alam, Jaakko Hollmén and Rahim Rahmani. IEEE 36th International Symposium on Computer-Based Medical Systems (CBMS), 2023, pp. 646-653.

Toward Multi Modality Language Model - implementation of GPT-4o/Project Astra

Explore the rich flavors of Indian desserts with TunedLlavaDelights. Utilizing the in Llava fine-tuning, our project unveils detailed nutritional profiles, taste notes, and optimal consumption times for beloved sweets. Dive into a fusion of AI innovation and culinary tradition

An all-new OS that orchestrates autonomous agents as workers to execute tasks.

Implementation of the model: "(MC-ViT)" from the paper: "Memory Consolidation Enables Long-Context Video Understanding"

A comprehensive review of techniques to address the missing-modality problem for medical images

Simple Implementation of a Transformer in the new framework MLX by Apple

Simple Implementation of TinyGPTV in super simple Zeta lego blocks

Final project for the course LT2318 Artificial Intelligence: Cognitive Systems. The project concerns multimodal hate speech detection in memes.

Queen's University - Data Mining (CISC 873)

Open source scripts to create large scale datasets with rich detail for multi-modal models

[IJHCS] An assistant prototype for breast cancer diagnosis prepared with a multimodality strategy. The work was published in the International Journal of Human-Computer Studies.

Official Implementation for Pre-CoFactv2 (AAAI-23 DeFactify2.0 Workshop 1st Place)

PyTorch Implementation of the paper "MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training"