Issue 69

Highlights

AI-powered visual search comes to the iPhone

Visual search is coming to the iPhone, powered by Apple Intelligence, Apple’s suite of AI capabilities. The Camera Control, the new button on the iPhone 16 and 16 Plus, can launch what Apple calls “visual intelligence” — basically a reverse image search combined with some text recognition.

Measurement is the key to helping keep AI on track

The creation of generative AI requires a new approach to evaluating, or measuring, the technology — one that combines technical and social aspects.

A fast and flexible approach to help doctors annotate medical scans

“ScribblePrompt” is an interactive AI framework that can efficiently highlight anatomical structures across different medical scans, assisting medical workers to delineate regions of interest and abnormalities.

David Pocock calls for election ban on AI deepfakes with fake videos of Albanese and Dutton

Spokesperson says government ‘considering the advice’ of Australian Electoral Commission on regulating AI use.

Paper of the week

Quantum reinforcement learning

We’re slowly seeing more Quantum computing research enter the mainstream AI field. For example, this paper combines Quantum computing principles with Reinforcement Learning to improve model performance. Instead of using a neural network to create an agent's policy, they’re using a quantum circuit. It’s crazy and worth a look.

Video

Episode 301 - GitHub models with Amy Boyd

Join us in this episode where we talk with Amy Boyd, Principal Cloud Advocate at Microsoft about GitHub models.

Smita Nachan - Humans in AI

Dive into the fascinating world of AI with Smita Nachan, a Microsoft MVP with over 14 years of expertise in M365 technology! Discover how AI is transforming workflows, making life a bit easier (and smarter) for us all. Smita shares her insights on the synergy between humans and AI, proving that, at the end of the day, it's humans who train the AI.

Copilot L33t Sp34k | Threat intelligence in the AI space

Sarah and Sherrod DiGrippo, Director of Threat Intelligence Strategy at Microsoft, discuss how AI is changing the threat intelligence landscape and how threat intelligence can be used with AI to enhance security.

Articles

Binary quantization in Azure AI Search: optimized storage and faster search

As organizations continue to harness the power of Generative AI for building Retrieval-Augmented Generation (RAG) applications and agents, the need for..

Meta Llama: Everything you need to know about the open generative AI model

Like other generative AI models, Llama can perform a range of different assistive tasks, like coding and answering basic math questions, as well as summarizing documents in eight languages.

GraphRAG auto-tuning provides rapid adaptation to new domains

GraphRAG uses large language models (LLMs) to create a comprehensive knowledge graph that details entities and their relationships from any collection of text documents. This graph enables GraphRAG to leverage the semantic structure of the data and generate responses to complex queries that require a broad understanding of the entire text. In previous blog posts, […]

The Evolution of Llama: From Llama 1 to Llama 3.1

A Comprehensive Guide to the Advancements and Innovations in the Family of Llama Models from Meta AI

The Future of AI: Exploring Multi-Agent AI Systems

AI agents are all the rage lately. With their ability to reason out complex tasks, AI agents can answer questions and take actions even without being explicitly programmed to do so. Lately, the concept of multi-agent systems is gaining traction for their ability to provide checks and balances. I recently wrote an example of a multi-agent system, in the form of a questionnaire-answering agent. I wrote it to automate the process of answering questions from RFPs or questionnaires, something my job commonly requires - but it also makes a great demo.

GenAI with Python: Coding Agents

Build a Data Scientist AI that can query db with SQL, analyze data with Python, write reports with HTML, and do Machine Learning (No GPU…

CUDA-Free Inference for LLMs

In this blog, we discuss the methods we used to achieve FP16 inference with popular LLM models such as Meta’s Llama3-8B and IBM’s Granite-8B Code, where 100% of the computation is performed using OpenAI’s Triton Language.

Benchmarking Hallucination Detection Methods in RAG

Here are the hallucination detection methods considered in our study, all based on using LLMs to evaluate a generated response: We will compare the hallucination detection methods stated above across 4 public Context-Question-Answer datasets spanning different RAG applications.

Code

Mathematics of love Optimizing a Dining-Room-Seating Arrangement for Weddings with Python

Solving the Restricted Quadratic Multi-Knapsack Problem (RQMKP) with mathematical programming and Python.