Metric-Driven Development of a RAG system

• Luxin Zhang
• 14 Min To Read
• 12 Jan, 2024
• • Software engineering

Prerequisites

• Basic knowledge of pros and cons of LLM APIs
• Preliminary knowledge of RAG structure

Introduction

Retrieval Augmented Generation (RAG) is commonly used to enhance the performance of LLM-based systems or incorporate domain-specific knowledge to improve the generated answers of the LLM. However, effectively evaluating and tracking the performance of a RAG system is not straightforward, posing a real challenge in the development and monitoring of such systems in a production environment.

How Open is Generative AI? Part 2

• Raphaël Semeteys Luxin Zhang
• 20 Min To Read
• 19 Dec, 2023
• • Artificial intelligence and data science

Embarking on the second and last part of our ‘Generative AI Openness’ series, we earlier established a straightforward framework to gauge the openness of Large Language Models (LLMs) and utilized it to explore LLM development and the positioning of key players. We noticed a trend towards increasingly restricted LLM artifacts for OpenAI and Google, contrasted with Meta’s more open approach.

How Open is Generative AI? Part 1

• Raphaël Semeteys Luxin Zhang
• 18 Min To Read
• 25 Oct, 2023
• • Artificial intelligence and data science

Welcome to this two-part series on Generative AI Openness, where we explore the history, current landscape, and potential future of open collaboration and proprietary control in the development of Language Language Models (LLMs). In this first part, we will delve into the importance of inspecting the openness of each component in the LLM training process, and how this can impact the potential limitations on the model use or reuse imposed by one or more of its components. In the second part of this series, we will explore the potential benefits and drawbacks of sharing Generative AIs openly for the collective advancement of society.

Worldline Labs is already in green AI

• Luxin Zhang Liyun He-Guelton Olivier Caelen Li Guo
• 8 Min To Read
• 06 Oct, 2022
• • Artificial intelligence and data science

Introduction

In 2021, the International Energy Agency (IEA) published a Net-Zero Emissions (NZE) Roadmap [1] by 2050 which outlines a scenario that the IEA has built to comply with the Paris climate agreement and, from this report, we know that we have an imperative to reduce urgently our energy consumption. For instance, according to the NZE, the total energy supply must fall by 7% between 2020 and 2030 and remain at around this level to 2050. The energy resource regenerates much more slowly than the needs and at the same time the world’s population will still grow. Using energy efficiently and reducing energy consumption becomes a key concern.