curius graph

all pages

showing 4651-4700 of 198841 pages (sorted by popularity)
Eliciting Latent Knowledge - Google Docs
[2601.21571] Shaping capabilities with token-level data filtering
[2601.19062] Who's in Charge? Disempowerment Patterns in Real-World LLM Usage
A Straussian reading of The Adolescence of Technology | Zhengdong
Intentionally Designing the Future of AI
In (highly contingent!) defense of interpretability-in-the-loop ML training — AI Alignment Forum
[2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMs
[2512.24601] Recursive Language Models
[2505.10831] Creating General User Models from Computer Use
College at Age Sixteen: What Is Intelligence For?
None of us are easy to be with - by hasif 💌
How AI assistance impacts the formation of coding skills \ Anthropic
Best Of Moltbook - by Scott Alexander - Astral Codex Ten
Life on Claude Nine - Igor Babuschkin
How to increase your surface area for luck - by Cate Hall
what makes a person interesting? - by emily north
Text is king - by Adam Mastroianni - Experimental History
Anthropic’s “Hot Mess” paper overstates its case (and the blog post is worse) — LessWrong
Chemical hygiene | karpathy
The Easing Blueprint
[2512.23675] End-to-End Test-Time Training for Long Context
Philosophy behind Claude's Constitution
it's time to start bullying young Palantir employees
[2501.16946] Gradual Disempowerment: Systemic Existential Risks from Incremental AI Development
Survival without dignity - Rudolf's Writing
You're overspending because you lack values
Insights on Crosscoder Model Diffing
Design Waterloo
stanford-steam-tunnels.pdf
Being creative requires taking risks - by Henrik Karlsson
[2512.18792] The Dead Salmons of AI Interpretability
From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence
A common type of time sink - by Alex - Cognition Café
Free Online PCB CAD Library | Ultra Librarian
Iris
What stops me from writing | Kat Huang
Gödel, Escher, Bach
hyam is designing...
Home - informal
turntrout.com/alignment-phd
Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Models
naturalhazard.xyz/ben_jess_sarah_starter_pack.html
Factory | Agent-Native Software Development
Afterimage
Public Work by Cosmos
Streisand effect
Shipmap.org | Visualisation of Global Cargo Ships | By Kiln and UCL
dreamofmachin.es/machine_prophecy.html
Galois - Specifications Don't Exist
Can we safely automate alignment research? - Joe Carlsmith