Muhammad Umair's picture

Muhammad Umair

umair894

·

AI & ML interests

AI Consultant | Engineer | Global Trainer |PhD Scholar |6+ Years of Experience: Specializing in building AI-powered SaaS solutions and delivering innovative AI, generative AI, and multimodal automation use cases. I've successfully led and contributed to dozens of projects, creating impactful proofs of concept (PoCs) that drive business value. With expertise in large language models (LLMs), computer vision, NLP, automation, and AI system design, I help companies unlock the power of AI to transform ideas into actionable solutions. As a global AI trainer, I empower organizations and individuals with the tools and knowledge to harness AI’s potential. Let’s build the future of AI together!

Recent Activity

reacted to singhsidhukuldeep's post with 👍 1 day ago

Excited to share groundbreaking research from @Baidu_Inc on enterprise information search! The team has developed EICopilot, a revolutionary agent-based solution that transforms how we explore enterprise data in large-scale knowledge graphs. >> Technical Innovation EICopilot leverages Large Language Models to interpret natural language queries and automatically generates Gremlin scripts for enterprise data exploration. The system processes hundreds of millions of nodes and billions of edges in real-time, handling complex enterprise relationships with remarkable precision. Key Technical Components: - Advanced data pre-processing pipeline that builds vector databases of representative queries - Novel query masking strategy that significantly improves intent recognition - Comprehensive reasoning pipeline combining Chain-of-Thought with In-context learning - Named Entity Recognition and Natural Language Processing Customization for precise entity matching - Schema Linking Module for efficient graph database query generation >> Performance Metrics The results are impressive - EICopilot achieves a syntax error rate as low as 10% and execution correctness up to 82.14%. The system handles 5000+ daily active users, demonstrating its robustness in real-world applications. >> Implementation Details The system uses Apache TinkerPop for graph database construction and employs sophisticated disambiguation processes, including anaphora resolution and entity retrieval. The architecture includes both offline and online phases, with continuous learning from user interactions to improve query accuracy. Kudos to the research team from Baidu Inc., South China University of Technology, and other collaborating institutions for this significant advancement in enterprise information retrieval technology.

published a Space 1 day ago

umair894/finetuningunsloth

liked a Space 2 days ago

deepseek-ai/Janus-Pro-7B

View all activity

Organizations

umair894's activity

upvoted a paper 28 days ago

A theory of appropriateness with applications to generative artificial intelligence

Paper • 2412.19010 • Published Dec 26, 2024 • 1

upvoted 5 papers about 1 month ago

VideoRefer Suite: Advancing Spatial-Temporal Object Understanding with Video LLM

Paper • 2501.00599 • Published Dec 31, 2024 • 41

LTX-Video: Realtime Video Latent Diffusion

Paper • 2501.00103 • Published Dec 30, 2024 • 41

2.5 Years in Class: A Multimodal Textbook for Vision-Language Pretraining

Paper • 2501.00958 • Published Jan 1 • 99

Next Token Prediction Towards Multimodal Intelligence: A Comprehensive Survey

Paper • 2412.18619 • Published Dec 16, 2024 • 54

1.58-bit FLUX

Paper • 2412.18653 • Published Dec 24, 2024 • 74

upvoted a paper about 2 months ago

InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions

Paper • 2412.09596 • Published Dec 12, 2024 • 94

upvoted 2 papers 2 months ago

VisionZip: Longer is Better but Not Necessary in Vision Language Models

Paper • 2412.04467 • Published Dec 5, 2024 • 106

Florence-VL: Enhancing Vision-Language Models with Generative Vision Encoder and Depth-Breadth Fusion

Paper • 2412.04424 • Published Dec 5, 2024 • 59

upvoted a paper 6 months ago

TableBench: A Comprehensive and Complex Benchmark for Table Question Answering

Paper • 2408.09174 • Published Aug 17, 2024 • 52

upvoted a collection 8 months ago

Qwen2

Qwen2 language models, including pretrained and instruction-tuned models of 5 sizes, including 0.5B, 1.5B, 7B, 57B-A14B, and 72B. • 39 items • Updated Nov 28, 2024 • 356

upvoted a paper 8 months ago

VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time

Paper • 2404.10667 • Published Apr 16, 2024 • 17

upvoted 4 collections 9 months ago

LLaVA++ (LLaMA-3 and Phi-3-Mini)

Extending Visual Capabilities of LLaVA with LLaMA-3 and Phi-3 • 11 items • Updated Jun 11, 2024 • 23

ZeroGPU Spaces

ZeroGPU Spaces made by the community • 17 items • Updated Jun 6, 2024 • 233

Phi-3

Phi-3 family of small language and multi-modal models. Language models are available in short- and long-context lengths. • 26 items • Updated 28 days ago • 551

Meta Llama 3

This collection hosts the transformers and original repos of the Meta Llama 3 and Llama Guard 2 releases • 5 items • Updated Dec 6, 2024 • 710

upvoted 3 collections 10 months ago

PDF Document / OCR Datasets

Document datasets with .pdf files that are usable with pixparse libraries and tools. • 2 items • Updated Mar 30, 2024 • 47

HF-curated models available on Workers AI

A collection of models curated with Hugging Face that can be run on Cloudflare's Workers AI serverless inference platform. • 15 items • Updated Apr 2, 2024 • 50

🎭 Avatars

The latest AI-powered technologies usher in a new era of realistic avatars! 🚀 • 70 items • Updated Dec 24, 2024 • 81

upvoted a paper 11 months ago

The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits

Paper • 2402.17764 • Published Feb 27, 2024 • 608