@edhenry@sigmoid.social. (@EdHenry_) header
@edhenry@sigmoid.social. (@EdHenry_) profile picture

Things I like : Mathematics, Machine Learning, Causality, Networks, and Philosophy.

Los Angeles, CA
Joined January 2011
Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)Photo by @edhenry@sigmoid.social. (@EdHenry_)
@edhenry@sigmoid.social. retweeted
One of the biggest misconceptions people have about intelligence is seeing it as some kind of unbounded scalar stat, like height. "Future AI will have 10,000 IQ", that sort of thing. Intelligence is a conversion ratio, with an optimality bound. Increasing intelligence is not so much like "making the tower taller", it's more like "making the ball rounder". At some point it's already pretty damn spherical and any improvement is marginal. Now of course smart humans aren't quite at the optimal bound yet on an individual level, and machines will have many advantages besides intelligence -- mostly the removal of biological bottlenecks: greater processing speed, unlimited working memory, unlimited memory with perfect recall... but these are mostly things humans can also access through externalized cognitive tools.
418
302
141
3,680
802,439
@edhenry@sigmoid.social. retweeted
Introducing RoboCap-10k. 10,000 hours of open egocentric data with metric hand / head pose on every frame. Captured on RoboCap across 11 countries and 21 industries. GroundedAPI produces 3D enrichment giving robot policies high-fidelity action grounding. with @BitRobotNetwork @huggingface @rerundotio ↓
14
15
10
111
21,364
@edhenry@sigmoid.social. retweeted
Why are old stories that were not even covered up all that well become sudden international and geopolitical news? The timing is suspicious, so worth asking: who benefits and why now? If Jensen, the premier champion of open models, is taken out then anyone who wants AI closed, centralized, and controlled might benefit. Why now? The midterms seem relevant. But also perhaps the upcoming IPOs.
Breaking: Nvidia’s board is preparing to oust CEO Jensen Huang after he had two extramarital affairs, including one with a suspected Chinese spy. on.wsj.com/4jcpr01
291
762
114
6,360
477,267
@edhenry@sigmoid.social. retweeted
"The trouble with definitions is that although they can be illuminating, witty, amusing, original and even revolutionary, there is one thing – and perhaps one thing only – which they cannot do: define a thing" George Mikes (English Humour For Beginners)
11
52
9
322
@edhenry@sigmoid.social. retweeted
Dear followers, please see this Substack from @MITshapingwork about my research with Andrea Manera and Pascual Retrepo on the US tax code's bias toward automation. How the US tax code incentivizes companies to replace workers with AI
The US tax system is biased against workers in favor of machines and algorithms, and has become more so in recent years. Our latest Substack post unpacks research from @DAcemogluMIT and co-authors: mitstonecenter.substack.com/…
5
49
2
190
23,008
@edhenry@sigmoid.social. retweeted
lol fuck that. If you are interested in something, dive deep. Write that essay, make the video, code the project, launch the company. The magic of tech is that you don’t need to “establish yourself” for a decade to have an opinion. You live only once motherfuckers, go for it.
This quoted post is unavailable.
1
5
73
3,199
@edhenry@sigmoid.social. retweeted
PyTorch 2.15 is a pretty big release if you run models on a Mac. Over the past months, the @PyTorch team at @huggingface has been working upstream on MPS, replacing MPSGraph paths with Metal kernels. Some of the results are kind of wild: - MoE decoding is now 5-6.5x faster than in PyTorch 2.14 - Qwen3.6-35B-A3B: 9.4 → 61.3 tok/s on an M3 Ultra - LFM2.5-8B-A1B: 23.6 → 123.6 tok/s - softmax over a 128k vocab is ~8x faster - FP8 is now supported on MPS - Wan2.2 VAE decoding goes from ~270 GiB peak driver memory to ~8 GiB This isn’t a specific runtime, or a separate inference engine, it’s just PyTorch getting faster on Apple silicon! Impressive work by @Is36E! 2.16 is already looking even faster 👀 huggingface.co/blog/mps-pyto…
4
12
2
62
2,253
@edhenry@sigmoid.social. retweeted
Today, @Meta is open sourcing the weights for Muse Spark Open, the most capable US open-weights model to date. It’s integrated with the PyTorch ecosystem: it ships with Helion kernels, and TorchTitan can be used to further tune the model. More in the blog: pytorch.org/blog/muse-spark-…
7
14
94
8,264
@edhenry@sigmoid.social. retweeted
The ideal outcome for science isn't a couple of labs using proprietary models and millions of dollars to crack trophy problems, it's to have millions of AI-aided human scientists who follow their curiosity and become way more ambitious
159
343
56
2,648
92,343
@edhenry@sigmoid.social. retweeted
10 remarks on AI and math that can be true all at once: 1. The recent progress is stunning and exciting. 2. The Quasi-Riemann Hypothesis is arguably Fields-medal level. 3. Mathematics is being transformed and there is no way back. 4. OpenAI doesn't give a fuck about the future of mathematics. 5. Half-baked unreadable preprints claiming major results can effectively stall mathematical progress. 6. Some mathematicians are responding with naive, overly defensive comments. 7. The social-media wankers framing this as "OpenAI tramples gatekeeping" have no experience of mathematical research. 8. The product of mathematics is clarity and understanding. Not theorems, by themselves. 9. Mathematics is a critical infrastructure of human cognition. The human need for understanding isn't obsolete and it can't be offloaded to machines. 10. We haven't "solved" mathematics and we won't get there before the Sun burns out.
158
465
58
2,964
160,013
@edhenry@sigmoid.social. retweeted
How AI shifted from a transductive paradigm (token-by-token answer completion) to an inductive paradigm (on-the-fly reasoning chain synthesis), in one chart. This is the big story of 2025 and 2026. This is what made coding agents work.
159
176
36
1,663
123,403
@edhenry@sigmoid.social. retweeted
Introducing RoboJEPA, the first scaling law for multiembodiment world models. We scaled JEPA predictors up to 8B params, showed scaling laws extrapolate and correlated compute scaling with planning performance. We release all checkpoints and the code. robojepa.github.io
38
139
10
810
57,547
@edhenry@sigmoid.social. retweeted
After a brief but institution-eroding slop science era, it increasingly looks like we are going to have two revolutions from AI: 1) Everything ever published will be re-read and re-judged in ways that human scientists never anticipated 2) Novel discoveries will start to come fast
156
220
35
2,778
233,386
@edhenry@sigmoid.social. retweeted
Five years ago this month, OpenAI released a dataset of grade school math problems, which all AIs at the time struggled with. Consider some of the problems:
7
87
9
987
104,414
@edhenry@sigmoid.social. retweeted
22
94
7
623
39,629
@edhenry@sigmoid.social. retweeted
Another reason why local AI is absolutely necessary.
192
267
86
4,920
367,155
@edhenry@sigmoid.social. retweeted
Today we're unveiling Trillium Labs @trillium_labs, a new non-profit to foster the open science of frontier AI. We're building open post-training recipes and will expand into open infra to study RSI, reward-hacking, multi-agent systems, and whatever comes next. We're built around the theory of change that you need more eyes to solve hard technical problems. We have faith in the scientific methods and communities that humanity has built, and worry that AI is becoming too closed to utilize them. Trilliums are wildflowers that bloom briefly in the spring, before the forest canopies fill out. Though they are small, they lay the foundation for the cycles of growth and nourishment through the rest of the year. At Trillium Labs, the recipes will be the slow nutrients for the seasons and the model releases will be the blooms. Building an institution dedicated to this is needed because, much as nature’s trilliums are slow to expand and grow, the open-ecosystem needs time and dedicated resources to catch up. I co-founded with with a long-time friend and collaborator Tom Zick (@thesezickbeats). We're hiring (full time + student collabs/interns), we're fundraising, and we're looking for compute. Please get in touch if you're interested in helping out. Offices based in the Bay Area and Cambridge MA, remote okay. I’m in the Bay Area until for The Curve and COLM to connect with people who are interested. We’re thankful to have initial support from Halcyon Futures and Schmidt Sciences with more funding en route to enable our ambitions of scaling. Our advisors @Thom_Wolf, @HannaHajishirzi, @gneubig and @ctnzr have been instrumental to building the ecosystem that exists today, and I’m stoked to get to keep working with them.
250
300
83
3,277
262,206
@edhenry@sigmoid.social. retweeted
In light of the latest Anthropic's latest not-particularly-oblique attempt to build a case against open weights, I thought I would write down how I currently think about open weight AI models. (1) Open weights have been deeply, irreplaceably useful for AI safety.
22
80
18
526
49,169
@edhenry@sigmoid.social. retweeted
緑が濃い場所、ただただ落ち着く🌿🍃
16
246
4
2,668
34,588
Jev is a large cross-section of communities less familiar with pre-transformer machine learning experiencing NFL for the first time. 😬
1
2
105
I should've included a paper link in the first post but I forgot 🤷‍♂️ so here it is: cs.ubc.ca/~hutter/earg/paper…
79