LLMs

VideoRAG: Redefining Long-Context Video Comprehension

Discover VideoRAG, a framework that fuses graph-based reasoning and multi-modal retrieval to enhance LLMs' ability to understand multi-hour videos efficiently.

Agentic AI, LLMs, RAGs, Video Analysis, Vision Language Models

Video-RAG: Training-Free Retrieval for Long-Video LVLMs

Learn how Video-RAG boosts training-free and low-compute long-video understanding by pairing OCR, ASR, and open-vocabulary detection with any long-video LVLMs.

RAGs, VLMs

Object Detection and Spatial Understanding with VLMs ft. Qwen2.5-VL

What if object detection wasn't just about drawing boxes, but about having a conversation with an image? Dive deep into the world of Vision Language Models (VLMs) and see how

Computer Vision, LLMs, NLP, Uncategorized, Vision Language Models, VLMs

Inside Sinusoidal Position Embeddings: A Sense of Order

In the groundbreaking 2017 paper “Attention Is All You Need”, Vaswani et al. introduced Sinusoidal Position Embeddings to help Transformers encode positional information, without recurrence or convolution. This elegant, non-learned

Language Models, LLMs, NLP

Inside RoPE: Rotary Magic into Position Embeddings

Self-attention, the beating heart of Transformer architectures, treats its input as an unordered set. That mathematical elegance is also a curse: without extra signals, the model has no idea which

Language Models, LLMs, NLP

FineTuning Gemma 3n for Medical VQA on ROCOv2

What if a radiologist facing a complex scan in the middle of the night could ask an AI assistant for a second opinion, right from their local workstation? This isn't

Computer Vision, Generative AI, Generative Models, LLMs, Multimodal Models, NLP, Transformer Neural Networks, Vision Language Models, Vision Transformer, VLMs