Rami's Readings #37
The latest on Chain-of-Thought Prompt Engineering, LLM Surveys, Adobe Generative AI, FTX Trials, Business Speed, Remittances, and more.
Welcome to Rami’s Readings #37 - a weekly digest of interesting articles, videos, Twitter threads from my various sources across the Internet. Expect a list of reads covering technology, business, culture, fashion, travel and more. Learn about what I do at ramisayar.com/about.
Upcoming Apéro & Intellect Gatherings!
Apéro & Intellect is a curated series of intimate gatherings discussing Artificial Intelligence designed to stimulate both your mind and your palate. Whether you're a coffee connoisseur, a tea enthusiast, or a cocktail aficionado, this is the place to be! Read the recap of #1. 🎟 For Subscribers Only.
🗓 Date: October 15th | ⏰ Time: 11am | 📍 Location: New York City
🗓 Date: October 19th | ⏰ Time: 8pm | 📍 Location: Montreal
🗓 Date: October 20th | ⏰ Time: 8pm | 📍 Location: Boston
🤖 AI Reads
Can You Run It? LLM Version
Notes: Handy tool to find out if your GPU can run or fine-tune a particular model.
Meta-CoT: Generalizable Chain-of-Thought Prompting in Mixed-Task Scenarios with Large Language Models
Notes: Neat! CoT continues to be a promising avenue for prompt engineering.
Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity
Notes: New survey preprint on ArXiv.
A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethics
Notes: Another great survey of LLMs for Healthcare.
InstructRetro: Instruction Tuning Post Retrieval-Augmented Pretraining
We hypothesize that pretraining with retrieval makes its decoder good at incorporating context for QA. Our results highlights the promising direction to obtain a better GPT decoder for QA through continued pretraining with retrieval before instruction tuning.
Adobe Is Working on Generative AI Video Manipulation
Notes: Adobe really is an AI-first company now.
A Tiny Startup Has Helped Intel Trounce AMD and Nvidia in Critical AI Tests — Is It Game over Already?
Notes: Skeptical about this specific startup but I will share again… Inference will be free in the long run. #Edge-AI
💼 Business Reads
You have to read
’s coverage of the FTX trials. My favorite was day four.
How Is It Still Getting Worse for Sam Bankman-Fried?
Notes: Entertaining read on the current legal proceedings against Bankman-Fried.
The Macro Impact of Short-Termism
Notes: Stop cutting your R&D budget when you have profit pressures.
At the macro level, short-termist distortions slow growth by 5 basis points yearly and lower social welfare by about 1%.
20 Years in Building Companies Taught Me: Speed Is King.
Notes: I agree with this take on speed. Fast experimentation is critical to learning and reaching product-market fit. Zuck is generally right, move fast and break things.
China’s Rich Entrust Total Strangers to Sneak Cash Out of the Country
Notes: Like inflation, I believe memories of export controls will continue to persist beyond their end date.
For individuals, even as Covid lockdowns are lifted, the memories of the extreme curbs remain.
That is all for this week. Signing off from Harlem, NYC.