momentmaker

health

wellnessfx - Improve health, fitness, wellness through blood tests, doctors, nutrition

driphydration - Drip Hydration Mobile IV Therapy | NAD, Dehydration, Hangover, Flu, Myers Cocktail

elysiumhealth - Breakthroughs in aging research are no longer a myth of science fiction but scientific fact. Elysium makes first-of-their-kind health products and technologies backed by partnerships with institutions at the forefront of aging research.

obasan - Obasan Mattress

veekaybee

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts of models in prod eagerly sought.

Foundational Concepts

Pre-Transformer Models

yoavg

Reinforcement Learning for Language Models

Yoav Goldberg, April 2023.

Why RL?

With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much

sikang99

Compact Language Models

Abbrs

• GPT : Generative Pretrained Transformer
• LLaMA : Large Language Model Meta AI, Meta
• GLM : Generic Language Model, Tsinghua University
• LoRA : Low-Rank Adaptation, MS
• LiGO : Linear Growth Operator, MIT
• PEFT : Parameter-Efficient Fine-Tuning, Hugging Face
• LoRA + DeepSpeed + CPU offloading

rain-1

Purpose

Bootstrap knowledge of LLMs ASAP. With a bias/focus to GPT.

Avoid being a link dump. Try to provide only valuable well tuned information.

Prelude

Neural network links before starting with transformers.

kariy

ZERO-KNOWLEDGE PROOF

🪅 General

[Recommended] Why and How zk-SNARK Works: Definitive Explanation

This paper aims to explain zk-SNARK construction in a simple and accessible way, using examples and answering common questions. Its contribution is a gradual increase in complexity to help readers understand without prior knowledge of cryptography or advanced math. The paper not only explains how zk-SNARK works, but also why it works and how it was developed.

1. Introduction to ZK
2. [MIT IAP 2023] Modern Zero Knowledge Cryptography
3. Curated resources for ZK stuff by Ingonyama

bilzard

質問

Psuedo labelingとは何ですか？

回答

Psuedo labelingは、深層学習のモデルを訓練するための手法の一つです。この方法では、予測が高い信頼性を持つデータポイントを、訓練データとしてモデルに使用します。この方法を使うことで、ラベル付けされていないデータも訓練に利用することができます。

例えば、あるタスクについてラベル付けされたデータセットがありますが、そのデータセットは非常に小さいため、モデルを訓練するには不十分です。このような場合、psuedo labelingを使うことで、より多くのデータをモデルに使用することができます。

ChristopherA

STATUS: RAW DRAFT

Goal is an opinionated version of something like Cryptographic Right Answers but focused on the needs of the decentralized identity and cryptocurrency ecosystes, and the needs of the Patrons of Blockchain Commons.

See also my replies to:

• https://lists.w3.org/Archives/Public/public-credentials/2022May/0048.html
• https://lists.w3.org/Archives/Public/public-credentials/2022Apr/0109.html

Encrypting Data

travisbrown

arthurgousset

💡 List of crypto reading lists

🙌 Please note, full credit goes to every original author! Each files includes a link to the source and a hat tip to the individual I found it through, if I didn't stumble over it myself.

📚 I'm replicating the content here to learn in public and for ease of reference. This is entirely for personal study and no commercial use is intendend. Please comment below if you'd like your content removed, I will take it out immediately!

Contents