biencoder

A genral RAG Search chatbot, with SoTA RAG techniques such as HyDE, Hybrid retrieval with BM25 + RRF and Cross encoder reranking. Evaluated on the BEIR scifact dataset and compared all the different pipelines i tried along the way

Cross-Encoder-with-Bi-Encoder를 활용한 WebPage 데모

ViSecRAG: A comprehensive Retrieval-Augmented Generation framework featuring semantic-aware data chunking, vector-based retrieval, and fine-tuned embedding models optimized for Vietnamese language security and information extraction tasks.

RAG Framework Implementation (Retriever/Reranker Fine-Tuning & Inference)

This repository contains the implementation of a novel, 3-phase Vision-Language Pre-training (VLP) pipeline designed for specialized medical imaging. By utilizing a Dual-Encoder architecture and Knowledge Distillation, this model learns to capture both global context and fine-grained pathological features.

A Semantic Search Engine over open-arxiv dump 1991-2026

Business entity resolution pipeline for Amazon ML Challenge 2026 (Neural Ninjas). A fine-tuned bi-encoder with ScaNN retrieval, a LightGBM candidate filter, four cross-encoder rerankers and stacked LightGBM matchers link records across three sources. Public leaderboard: 0.9873.

Repository for the paper - Exploring Argument Retrieval for Controversial Questions Using Retrieve and Re-rank Pipelines

Dense Retrieval with Low-Rank Adaptation for Scalable Multi-Tenant Search