RAFT: Adapting Language Model to Domain Specific RAG

Hong-Wei Wu published on 2024-10-31 included in category Paper Introduction

RAFT paper analysis: Learn how to train LLMs for Domain-Specific RAG, combining external documents with internal knowledge to boost specialized domain QA performance.

REPLUG: Retrieval-Augmented Black-Box Language Models

Hong-Wei Wu published on 2024-10-31 included in category Paper Introduction

Explore how Retrieval-Augmented Generation (RAG) enhances black-box LLMs. This article details the NAACL 2024 paper REPLUG, discussing its innovative methods for Inference and Training stages to improve LLM answer quality and effectively reduce hallucination.

Python's Small Integer Cache: Faster Code with a Hidden Trick

Hong-Wei Wu published on 2024-10-03 included in category Python Tutorial

Discover Python's Small Integer Cache! Learn how Python optimizes common integers (-5 to 256) for faster performance and efficient memory use. Simple examples included.

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

Hong-Wei Wu published on 2024-07-29 included in category Paper Introduction

Explore Cambrian-1, NYU's deep dive into vision-centric Vision-Language Models (VLMs). Discover key insights on visual encoders, connector design (SVA), training strategies, and new open-source tools like CV-Bench & Cambrian-7M data to advance AI's visual understanding.

Better & Faster Large Language Models via Multi-token Prediction

Hong-Wei Wu published on 2024-07-18 included in category Paper Introduction

Explore Meta AI's groundbreaking Multi-Token Prediction Model. This deep dive explains how predicting multiple tokens at once can enhance LLM performance, detailing its unique architecture and clever techniques for reducing GPU memory usage. A must-read for AI and ML enthusiasts.

Tuning LayerNorm in Attention: Towards Efficient Multi-Modal LLM Finetuning

Hong-Wei Wu published on 2024-07-08 included in category Paper Introduction

Discover how to efficiently train powerful Multimodal LLMs (MLLMs). This post explores a new ICLR 2024 technique that achieves top performance, rivaling full fine-tuning, by simply adjusting the LayerNorm in attention blocks—all while saving significant GPU memory.

GAIA: A Benchmark for General AI Assistants

Hong-Wei Wu published on 2024-06-27 included in category Paper Introduction

Why do powerful AIs like GPT-4 struggle with tasks humans find easy? Dive into GAIA, the game-changing benchmark from Yann LeCun's team, and discover how it redefines the true capabilities of a General AI Assistant beyond standard tests. A must-read for anyone in AI.