mpnet

Scripts and Approach for Amazon ML Challenge

Lightweight text and scans processing: embedding, document processing, OCR, OMR, etc, with inference via ggml in pure C++

This repo shows how foundational model hallucinates and how we can fix such hallucinations using fine-tuning them

A python project to generate a learning path on a specific topic or question.

A Python tool for finding matching segments in video transcripts using semantic similarity. and a simple txt file

An experiment driven Search Engine Project, developed to index and retrieve best documents given a query using ensemble of models.

Clusters various mechanical problems based on their characteristics.

Raw Torch, heavily annotated, pretrainable MPNet

This repository contains the code for submission made at SemEval 2022 Task 5: MAMI

Advanced NLP project where we needed to build a Sarcasm/Irony classifier. It has many methods like BiLSTM, Transformers to BERT/ T5 / MPNET finetuning

A state-of-the-art Retrieval-Augmented Generation (RAG) system with hybrid search, multi-hop reasoning, answer verification, and source citation — delivering accurate, trustworthy, and context-aware answers from large document collections.

An end-to-end theme analysis pipeline for wellbeing data, combining AI-based classification with unsupervised topic discovery to power diagnostics, dashboards, and insight generation.

FastAPI ML microservice that detects Merchant Cash Advance (MCA) and line-of-credit lender payments in bank transactions using all-mpnet-base-v2 sentence-transformer embeddings, keyword, and fuzzy matching.

Fully offline hybrid RAG concierge for luxury hotels with confidence aware query routing, FAISS HNSW semantic search, local Snoozy 13B LLM fallback, multi layer caching and live metrics. No cloud. No data leakage.

Detect and analyze resource leaks in Android apps using method signature matching and NLP-based similarity detection.