curius graph

all pages

showing 36101-36150 of 160880 pages (sorted by popularity)
o1: A Technical Primer — LessWrong
LatentQA: Teaching LLMs to Decode Activations Into Natural Language
Why Don't We Just... Shoggoth+Face+Paraphraser? — LessWrong
Alignment faking in large language models - YouTube
No hype, just works: How Comma reached 100M miles in autonomous driving | E2011 - YouTube
Why Not Just Outsource Alignment Research To An AI? — LessWrong
Tips and Code for Empirical Research Workflows — AI Alignment Forum
Recipe: Hessian eigenvector computation for PyTorch models — LessWrong
Coercion is an adaptation to scarcity; trust is an adaptation to abundance — LessWrong
Agency begets agency — LessWrong
Preventing model exfiltration with upload limits
Diffusion language models – Sander Dieleman
[2211.15089] Continuous diffusion for categorical data
Spying on Python with py-spy
LLMs are (mostly) not helped by filler tokens — LessWrong
Win/continue/lose scenarios and execute/replace/audit protocols — AI Alignment Forum
RL with KL penalties is better seen as Bayesian inference — LessWrong
Misalignment and Strategic Underperformance: An Analysis of Sandbagging and Exploration Hacking
Decoding opaque reasoning in current models (empirical project proposal) [public] - Google Docs
Natural Latents: The Math — LessWrong
Teaching ML to answer questions honestly instead of predicting human answers | by Paul Christiano | AI Alignment
Unlearning via RMU is mostly shallow — AI Alignment Forum
When does training a model change its goals?
Bogong moths use a stellar compass for long-distance navigation at night | Nature
Data representation – CS 61
Fusion energy start-up claims to have cracked alchemy
If a tree falls on Sleeping Beauty... — LessWrong
Victor Weisskopf - Wikipedia
David McCullough - Wikipedia
Herman Feshbach - Wikipedia
Simplex Progress Report - July 2025 — LessWrong
Suddenly, Trait-Based Embryo Selection - by Scott Alexander
I've tracked my last 800 flights. Here's when you really need to get to the airport.
Why future AI agents will be trained to work together
An Intuitive Explanation of Quantum Mechanics — LessWrong
Configurations and Amplitude — LessWrong
Opening the Black Box: Interpretable LLMs via Semantic Resonance Architecture
Evolving Language Models without Labels: Majority Drives Selection, Novelty Promotes Variation
LLM-JEPA: Large Language Models Meet Joint Embedding Predictive Architectures
A Toy Model of Mechanistic (Un)Faithfulness
Synthesizing Standalone World-Models (+ Bounties, Seeking Funding) — AI Alignment Forum
Artusi: Lesson 8: Triads
How Well Does RL Scale? — Toby Ord
How to Create Synthetic Data at High Quality for Fine-Tuning LLMs
Introducing RFM-1: Giving robots human-like reasoning capabilities
ARC progress update: Competing with sampling — LessWrong
[2511.01836] Priors in Time: Missing Inductive Biases for Language Model Interpretability
[2511.00617] Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
Circuits Updates - April 2024
Watermarking of Large Language Models - YouTube