Mayank Laddha articles - 13 total
Sports Highlights Generation Using Transcript and LLM
I wanted to try and see if I could use gemini to generate highlights. This is what the code...
GRPO to get structured data
I saw this line "GRPO gives developers a way to teach models how to reason better, faster, and...
My take on the Agentic Object Detection
Here are the steps: Segmenting Everything with SAM : We detect everything and worry about filtering...
Semantic search on top of object detection
Semantic search on top of object detection: Imagine this across multiple cameras with tracking,VQA...
Food Recognition and Nutrition Estimation using OpenAI
Here is how you can build a simple food recognition and nutrition estimation app using OpenAI in just...
Your Personal AI Cricket Coach
Your Personal AI Cricket Coach Get Feedback Anytime Anywhere: AI can make learning more accessible,...
A multi-head classifier using SetFit for query preprocessing
Query preprocessing is a crucial step in customer-centric applications, ensuring accurate routing and...
Faiss with sqlite for RAG
Want to use faiss for Local RAG? Okay, but where to store my chunks(metadata). Solution: Connect...
Detecting Hallucinations in LLMs with Discrete Semantic Entropy and Perplexity
When working with large language models (LLMs), spotting hallucinations can be tricky. Instead of...
Simple Code to Understand Self-Reflection (Agentic Design Pattern)
Reflection/self-reflection is a bit underrated. If your application relies on prompting, I strongly...
Using DSPy(COPRO) to refine prompt instructions
I recently tested DSPy with phi3, aiming to use COPRO to refine the prompt instructions and achieve...
An easy way to remove PII before sending to LLMs
Not all scenarios demand perfect anonymization. In less critical cases, a lightweight anonymization...
Graph + LLM or simply LLM for summarization?
let's say I have some documents. I want to generate a summary (get an overall essence) may be based...