Sources used in Workshop
Introduction
Welcome
I-Ching Exercise
Concepts
Compound AI
- The Shift from Models to Compound AI Systems
- Sober AI is the Norm
- The Stickle-Brick Approach to Big AI
Types of AI and ML Relevant to Libraries
Naïve Bayes
- Automated Classification to Improve the Efficiency of Weeding library Collections
- Bayes Theorem
- Naïve Bayes
Natural Language Processing
Basics of Neural Networks
- Machine Learning for Beginners: An Introduction to Neural Networks
- [Neural Network (machine learning)](https://en.wikipedia.org/wiki/Neural_network_(machine_learning)
- What is a neural network?
Transformers
- Attention Is All You Need
- Transformer Explainer Demonstration
- Transformer Explainer: Interactive Learning of Text-Generative Models
Ethical Considerations
Bias
- Evaluating and Mitigating Discrimination in Language Model Decisions
- Gemini for Google Cloud and responsible AI
- How should AI systems behave, and who should decide?
- Managing Bias in AI
- Shedding light on AI bias with real world examples
- There’s More to AI Bias Than Biased Data, NIST Report Highlights
Academic Fraud
- Artificial Intelligence (AI) policies for Nature Publishing
- Crafting AI-Complementary Skills and Bulletproof Assessments
- Generative AI Can Harm Learning
- Guidelines for Artificial Intelligence (AI)-Generated Text for IEEE
- How Much Research Is Being Written by Large Language Models?
- Mapping the Increasing Use of LLMs in Scientific Papers
- Science Journals Editorial Policies
- The use of generative AI and AI-assisted technologies in writing for Elsevier
- X Post on use of LLMs in an Elsevier article
Creator Attribution and Copyright
- Generative AI Has a Visual Plagiarism Problem Experiments with Midjourney and DALL-E 3 show a copyright minefield
- New York Times Legal Complaint December 2023
- Training Generative AI Models on Copyrighted Works Is Fair Use
Privacy Concerns in Large Language Models
- Generative AI's privacy problem
- Privacy in Large Language Models: Attacks, Defenses and Future Directions
- Scalable Extraction of Training Data from (Production) Language Models
Guidelines for Incorporating AI
- ARL: Research Libraries Guiding Principles for Artificial Intelligence
- European Union Requirements for Trustworthy AI
- Ontario Council of University Libraries: Artificial Intelligence/Machine Learning Final Report and Strategy
Carbon Footprint of LLMs
- AI Chatbots: Energy usage of 2023’s most popular chatbots (so far)
- Assessing the environmental impact of large language models
- Gen AI’s Environmental Ledger: A Closer Look at the Carbon Footprint of ChatGPT
- Google Environmental Report 2024
- LLMCarbon: Modeling the end-to-end Carbon Footprint of Large Language Models
- Microsoft How can we advance sustainability? 2024 Environmental Sustainability Report
- Reducing the Environmental Impact of Generative AI: a Guide for Practitioners
Deepfakes
- Deepfake appears to show Bangladeshi politician dropping out of race
- Deepfake appears to show Eminem endorsing South African opposition party
- Deepfake of US congressman appears to support Taiwan’s Democratic Progressive Party
- Deepfake shows Ukrainian president Volodymry Zelensky dancing
- Fake image of Donald Trump with Black Voters
- Fake Joe Biden robocalls
- Fake recording of Mexico City Mayor interfering with Elections
- Fake Views: : The Good, The Bad, And The Ugly Of Soviet Photoshopping
- Series of deepfakes encourage election boycotts in Pakistan
- How AI is resurrecting dead Indian politicians as election looms
- Wikipedia: Deepfake
AI Slop
- Google Searches Prefer AI Spam to Real Content
- Slop is the new name for unwanted AI-generated content
- The death (again) of the internet as we know it
- Too big to care
- Why I redesigned my home page?
Hallucinations and Generative AI
- Andrej Karpathy X post on 8 December 2023
- A.I. Has a Measurement Problem
- Best Strategies to Minimize Hallucinations in LLMs: A Comprehensive Guide
- Hallucinations, Errors, and Dreams
p(doom) and Artificial General Intelligence (AGI)
- Artificial General Intelligence
- Benefits & Risks of Artificial Intelligence
- Ethical Issues in Advanced Artificial Intelligence
- List of p(doom) values
- P(doom) is AI’s latest apocalypse metric. Here’s how to calculate your score
- State of AI Engineering 2023
Exploring Large Language Models
OpenAI's ChatGPT
- Arizona State University
- ChatGPT
- ChatGPT API
- DALL-E 3
- Harvey
- Microsoft Copilot
- Open AI
- Open AI GPTs.
- OSCAR
Anthropic's Claude
Google Gemini
- Download or install libraries to access Gemini
- Google Gemini
- Gemini 1.0 Pro
- Gemini 1.5 Flash
- Gemini 1.5 Pro
- Why Google’s AI tool was slammed for showing images of people of colour
Local LLaMa
Generative AI Images and Videos
- Discord
- Dream Machine
- Google's Gemini
- Luma Labs
- Open AI
- Sora
- Why Google’s AI tool was slammed for showing images of people of colour
Prompt Engineering
- Antropic Prompt Engineering Guide
- Build with Claude Tool use (function calling)
- Function calling for Llama 3.1 models!
- Intro to Function Calling with Gemini API
- Open AI's Function Calling Guide
- Principled Instructions Are All You Need for Questioning LLaMA-1/2, GPT-3.5/4
- Prompt Engineering Guide
Retrieval Augmented Generation (RAG)
- A Simple Guide To Retrieval Augmented Generation Language Models
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Training Large Language Models (LLMs)
- Fine Tune Claude 3 Haiku
- Guide for Fine-tuning
- GGUF
- gpt4all
- HuggingFace
- LLaMA from AI Meta
- LLaMA.cpp
- LLaMA.cpp with Docker
- LLaMA.cpp Python SDK
- LoRA: Low-Rank Adaptation of Large Language Models
- loralib
- mlx-lm
- mlx-ml LoRA documentation
- Open AI
- Open AI - Files API
- OpenAI Fine-Tune User Interface
- Pytorch
- Unslouth
- What is low-rank adaptation (LoRA)?