noahdgoodman (@noahdgoodman) / X

noahdgoodman

341 posts

noahdgoodman

@noahdgoodman

humans& co-founder. professor of natural and artificial intelligence @Stanford. (@StanfordNLP @StanfordAILab) ex: Google DeepMind, UberAI

Joined November 2019

Pinned
noahdgoodman
@noahdgoodman
Jan 20
I’ve co-founded humans& with @ericzelikman @TheAndiPenguin @gharik & @YuchenHe07 — a new AI lab working to bring people together, empower the better angels of our nature, and create the most optimistic future. At a time of tumult and danger, humanity needs a humanist AI.
humans&
@humansand
Jan 20
Today we introduce humans&, a human-centric frontier AI lab. We believe AI can be reimagined, centering around people and their relationships with each other. At its best, AI should serve as a deeper connective tissue that strengthens organizations and communities
60K
noahdgoodman
@noahdgoodman
Jan 29, 2025
Congrats to OAI on producing a reasoning model! Their opaque tweets demonstrate that they’ve (independently) found some of the core ideas that we did on our way to STaR.
Mark Chen
@markchen90
Jan 28, 2025
Congrats to DeepSeek on producing an o1-level reasoning model! Their research paper demonstrates that they’ve independently found some of the core ideas that we did on our way to o1.
424K
noahdgoodman
@noahdgoodman
Apr 8, 2024
When I first saw Tree of Thoughts, I asked myself: If language models can reason better by searching, why don't they do it themselves during Chain of Thought? Some possible answers (and a new paper): 🧵
110K
noahdgoodman
@noahdgoodman
Mar 24, 2024
This seems like a good time to mention that I've taken a part-time role at @GoogleDeepMind working on AI Safety and Alignment!
Anca Dragan
@ancadianadragan
Mar 22, 2024
So excited and so very humbled to be stepping in to head AI Safety and Alignment at @GoogleDeepMind. Lots of work ahead, both for present-day issues and for extreme risks in anticipation of capabilities advancing.
57K
noahdgoodman
@noahdgoodman
Jan 29, 2025
Replying to @tomlikestocode
I think it means in the future we will develop models by having smart and intense researchers stand in the shoulders of previous smart and intense researchers
15K
noahdgoodman
@noahdgoodman
Nov 1, 2025
we’re bending the arc of the future toward optimistic. help us build?
Eric Zelikman
@ericzelikman
Nov 1, 2025
we think humanity’s biggest challenges won’t be solved by ai thinking for 1000 hours coming back with an answer they’ll be solved by many collaborating humans, and ai that understands them and their different skills, goals, values, etc to empower them to do more together
42K
noahdgoodman
@noahdgoodman
Sep 19, 2024
You give your partner flowers. They become angry. Why? It turns out language models are good at this — superhuman affective cognition: arxiv.org/abs/2409.11733
25K
noahdgoodman
@noahdgoodman
Dec 27, 2023
Hey twitter! @mcxfrank and I are teaching a seminar "Topics in Natural and Artificial Intelligence", where we'll be thinking about the relations between cognitive psych and modern AI (mostly LLMs). What papers are we missing?
docs.google.com
Psych 247 Syllabus (Winter 2024)
Topics in Natural and Artificial Intelligence (Psych 247) Instructors: Noah Goodman and Mike Frank Description: We will read and discuss a selection of recent papers at the intersection of psychology...
21K
noahdgoodman
@noahdgoodman
Dec 9, 2023
What nobody is mentioning: state-space models (mamba, stripey hyena, etc) are a huge win for AI safety. The strict bottleneck between past and future provides a clean target for interpretation and intervention.
21K
noahdgoodman
@noahdgoodman
Jun 30, 2025
It turns out that a lot of the most interesting behavior of LLMs can be explained without knowing anything about architecture or learning algorithms. Here we predict the rise (and fall) of in-context learning using hierarchical Bayesian methods.
Ekdeep Singh Lubana
@EkdeepL
Jun 28, 2025
🚨New paper! We know models learn distinct in-context learning strategies, but *why*? Why generalize instead of memorize to lower loss? And why is generalization transient? Our work explains this & *predicts Transformer behavior throughout training* without its weights! 🧵 1/
GIF
19K
noahdgoodman
@noahdgoodman
Oct 31, 2024
New work with the awesome @aryaman2020 ! This started as some musings about “model organisms” for studying LM alignment, and ended up with really cool connections between ICL, Bayes, and alignment techniques.
Aryaman Arora
@aryaman2020
Oct 31, 2024
New paper! 🫡 In-context learning (ICL) is when LLMs infer how to do a task from examples. We know that the relationship between # of ICL examples and task accuracy is predictable. Can we predict the shape of the ICL curve using Bayesian assumptions? Our paper shows yes!
20K
noahdgoodman
@noahdgoodman
Mar 15, 2024
I’m thrilled for this to be public! The culmination of years of thinking about thinking.
Eric Zelikman
@ericzelikman
Mar 15, 2024
Language models today are trained to reason either 1) generally, imitating online reasoning data or 2) narrowly, self-teaching on their own solutions to specific tasks Can LMs teach themselves to reason generally?🌟Introducing Quiet-STaR, self-teaching via internal monologue!🧵
GIF
19K
noahdgoodman
@noahdgoodman
Nov 19, 2023
💚 (i don't work at openai, or know @sama, or have really any opinion about any of this... but all the cool kids are heartreposting and i fomo)
Sam Altman
@sama
Nov 19, 2023
i love the openai team so much
25K
noahdgoodman
@noahdgoodman
Nov 24, 2023
Y’all know what the * in Q* stands for, don’t you? Self TAught Reasoner.
arxiv.org
STaR: Bootstrapping Reasoning With Reasoning
Generating step-by-step "chain-of-thought" rationales improves language model performance on complex reasoning tasks like mathematics or commonsense question-answering. However, inducing language...
12K