Graham @_GR4HAM

@[email protected] Joined June 2009

Tweets

532
Followers

68
Following

1K
Likes

4K

Sakana AI @SakanaAILabs

2 weeks ago

What if we could evolve AI models like organisms in nature, letting them compete, mate, and combine their strengths to produce ever-fitter offspring? Excited to share our new work: “Competition and Attraction Improve Model Fusion” presented at GECCO’25🦎 where it was a runner-up…

38 154 824 170K 499

Download Image

Brave @brave

3 weeks ago

AI agents that can browse the Web and perform tasks on your behalf have incredible potential but also introduce new security risks. We recently found, and disclosed, a concerning flaw in Perplexity's Comet browser that put users' accounts and other sensitive info in danger.

93 585 4K 1.5M 1K

Download Image

Jiashuo Liu @liujiashuo77

3 weeks ago

We built FutureX, the world’s first live benchmark for real future prediction — politics, economy, culture, sports, etc. Among 23 AI agents, #Grok4 ranked #1 🏆 Elon didn’t lie. @elonmusk your model sees further 🚀🍀 LeaderBoard: futurex-ai.github.io

228 213 1K 7.4M 551

Download Image

Shai Shalev-Shwartz @shai_s_shwartz

4 weeks ago

Are frontier AI models really capable of “PhD-level” reasoning? To answer this question, we introduce FormulaOne, a new reasoning benchmark of expert-level Dynamic Programming problems. We have curated a benchmark consisting of three tiers, in increasing complexity, which we call…

103 411 4K 690K 2K

Download Image

Jack Morris @jxmnop

4 weeks ago

OpenAI hasn’t open-sourced a base model since GPT-2 in 2019. they recently released GPT-OSS, which is reasoning-only... or is it? turns out that underneath the surface, there is still a strong base model. so we extracted it. introducing gpt-oss-20b-base 🧵

159 470 6K 919K 4K

Download Image

Zack Witten @zswitten

2 months ago

Zack Witten @zswitten

2 months ago

https://t.co/VqkRK8aaYJ

9 29 1K 323K 333

Download Image

24 283 5K 723K 963

Download Image

Yiping Lu @2prime_PKU

2 months ago

Anyone knows adam?

271 465 5K 615K 505

Download Image

Mehul Damani @MehulDamani2

2 months ago

🚨New Paper!🚨 We trained reasoning LLMs to reason about what they don't know. o1-style reasoning training improves accuracy but produces overconfident models that hallucinate more. Meet RLCR: a simple RL method that trains LLMs to reason and reflect on their uncertainty --…

13 266 896 94K 612

Download Image

Midnight Maniac Sri @sridatta

2 months ago

'water is transparent only within a very narrow band of the electromagnetic spectrum, so living organisms evolved sensitivity to that band, and that's what we now call "visible light". ' (found via HN)

392 4K 46K 10.1M 13K

Download Image

Keyon Vafa @keyonV

2 months ago

Can an AI model predict perfectly and still have a terrible world model? What would that even mean? Our new ICML paper formalizes these questions One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵

211 1K 7K 1.4M 5K

Download Video

Machine Learning Street Talk @MLStreetTalk

2 months ago

AI is so smart, why are its internals 'spaghetti'? We spoke with @kenneth0stanley and @akarshkumar0101 (MIT) about their new paper: Questioning Representational Optimism in Deep Learning: The Fractured Entangled Representation Hypothesis. Co-authors: @jeffclune @joelbot3000

18 65 280 82K 178

Download Video

Fenil Doshi @fenildoshi009

2 months ago

🧵 What if two images have the same local parts but represent different global shapes purely through part arrangement? Humans can spot the difference instantly! The question is can vision models do the same? 1/15

5 114 593 62K 419

Download Gif

Paras Chopra @paraschopra

2 months ago

@jxmnop This paper has similar vibes - Train on cellular automata data - Freeze network and train last layer on chess move prediction - Voila! It works My intuition is that any data with sufficient number of patterns helps transformer build internal abstractions which later become easy…

5 6 100 6K 72

Download Image

noordung @n00rdung

2 months ago

Actually, no; in fact, quite contrary: the impact of "early massive galaxy formation" helps resolve what used to be a discrepancy between the general model & the observed cosmic microwave background.

cascadian realism fan 🌲 @realism_fan

2 months ago

Actually, no; in fact, quite contrary: the impact of "early massive galaxy formation" helps resolve what used to be a discrepancy between the general model & the observed cosmic microwave background.

1K 2K 19K 2.4M 10K

Download Image

5 31 255 14K 86

Davis Blalock @davisblalock

2 months ago

Deep learning training is a mathematical dumpster fire. But it turns out that if you *fix* the math, everything kinda just works…fp8 training, hyperparameter transfer, training stability, and more. [1/n]

15 150 1K 187K 2K

Download Image

Dr. Dominic Ng @DrDominicNg

2 months ago

Microsoft claims their new AI framework diagnoses 4x better than doctors. I'm a medical doctor and I actually read the paper. Here's my perspective on why this is both impressive AND misleading ... 🧵

278 1K 9K 1.6M 6K

Download Image

Luke Darlow @LearningLukeD

2 months ago

Learning where to look, learning what to see. Work on Continuous Thought Machines continues to be exciting! As a teaser, take a look at our approach to "Foveated Active Vision". Check it out 👇 (details in the thread) pub.sakana.ai/ctm

18 129 871 61K 533

Download Video

David Hall @dlwh

2 months ago

So about a month ago, Percy posted a version of this plot of our Marin 32B pretraining run. We got a lot of feedback, both public and private, that the spikes were bad. (This is a thread about how we fixed the spikes. Bear with me. )

Percy Liang @percyliang

4 months ago

9 17 304 313K 91

Download Image

23 102 1K 296K 1K

Download Image

Richard Socher @RichardSocher

3 months ago

If you studied algorithms, I'm sure you've heard of Dijkstra’s algorithm to find the shortest paths between nodes in a weighted graph. Super useful in scenarios such as road networks, where it can determine the shortest route from a starting point to various destinations. It's…

23 130 1K 140K 961

Download Image

Stella Li @StellaLisy

3 months ago

Spurious Rewards was not all‼️We now present spurious PROMPTS🤔 check out our latest findings and discussion on evaluation: tinyurl.com/spurious-prompt. Who knew Lorem ipsum can bring 19.4% gains compared to default prompt👀 Also, arXiv is out🤩 arxiv.org/abs/2506.10947📄