// curated_library

Seminal papers, key talks, and essential videos from the people shaping AI — filtered for signal, not noise.

paper

Claude's Model Spec (Constitutional AI)

Dario Amodei — 2024

The specification document that defines how Claude is trained to be helpful, harmless, and honest. A remarkable piece of AI alignment engineering made public.

alignment RLHF Constitutional AI safety
Open ↗
paper

The Forward-Forward Algorithm (2022)

Geoffrey Hinton — 2022

Hinton's alternative to backpropagation — training neural networks using only local, positive and negative data without ever computing a global gradient. Potentially significant for neuromorphic and on-device learning.

backpropagation learning algorithms neuromorphic alternative training
Open ↗
paper

Attention Is All You Need

Vaswani et al. (Google Brain) — 2017

The paper that changed everything. The transformer architecture introduced here is the foundation of every major language model in existence. If you work in AI, you must have read this.

transformer attention NLP foundational seminal
Open ↗
paper

Scaling Laws for Neural Language Models

Kaplan et al. (OpenAI) — 2020

The empirical laws governing how model performance scales with compute, data, and parameters. The theoretical backbone behind the scaling hypothesis that has driven frontier AI for five years.

scaling laws language models compute empirical ML
Open ↗
paper

Constitutional AI: Harmlessness from AI Feedback

Bai et al. (Anthropic) — 2022

The research paper behind Claude's training methodology. Describes how a set of principles ('constitution') guides RLHF without requiring human labellers for every harmful output.

Constitutional AI RLHF alignment Anthropic safety
Open ↗
paper

Deep Learning (textbook)

Goodfellow, Bengio, Courville — 2016

The canonical deep learning textbook — freely available online. Chapter 6 onwards remains the most rigorous treatment of feedforward nets, regularisation, and optimisation available.

textbook deep learning foundational Bengio
Open ↗