visiting researcher @princeton working with synthetic data for 3D computer vision

Lukas retweeted
skill issue
Replying to @Rainmaker1973
The actual number of moves that are not utterly stupid in chess is tiny and chess will be fully solved one day
194
2,190
53,088
2,436,936
Lukas retweeted
I'm so excited that our @theworldlabs team has achieved a major milestone today! Introducing Atlas - a first of its kind multimodal world model trained from scratch! 🚀 Atlas is capable of generating frames with pixel-perfect camera control, reconstructing large scenes from as few as one single input image, simulating space-time by reframing videos, natively outputting 3D spaces from one or more input images, composing multiple posed images into a consistent 3d world, and more! This is the best camera conditioned world model ever, opening doors to many possible use cases from VFX to robotics. I'm so so so proud of our team!♥️
Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.
326
1,102
9,731
1,261,966
Announcing our $130M Series A to build the Open Superintelligence Stack Led by Radical Ventures, with NVIDIA, Intel Capital, Dell Capital, and existing investors Train, deploy, and continuously improve your own models using our stack. Own your intelligence.
333
549
4,997
1,451,007
This makes me doubt the 90%+ success rate that every VLA-based policies claim I truly respect all the work behind public demos, and especially from hardware companies But from what I saw at #ICRA , even pick-and-place is still challenging for VLAs and requires human-in-the-loop
16
14
120
21,828
Lukas retweeted
When Opus 4.5 came out, it was a one-way door to a new way of engineering. Agents now do most of our coding. Knowing the inherent flaws and over-confidence of LLMs, we sent a clear message to our teams. Vibing and mission-critical infrastructure don’t go together. We’re sharing some of our early internal guidance in how we’re “agenting responsibly”, prioritizing security, durability, and availability at all times. vercel.com/blog/agent-respon…
79
157
1,830
216,656
ML research is showing it’s easier to replicate the human brain than thumbs Thank you for your attention to this matter
1
24
My plan B is to become a python-blender 3D artist
3
132
intellect-3 is here. this model has been my life the last 2 months and fucked my sleep schedule more than once. a huge team effort that im very proud of. some technical details:
Introducing INTELLECT-3: Scaling RL to a 100B+ MoE model on our end-to-end stack Achieving state-of-the-art performance for its size across math, code and reasoning Built using the same tools we put in your hands, from environments & evals, RL frameworks, sandboxes & more
12
13
184
17,145
I am so ugly in 3D
1
2
92
also it's a single architecture that predicts depth, segmentation and normals in one go. how neat
1
1
87
live feed from video cam was about 0.7 fps
1
75
It's diabolical that my mood each day for the whole day is determined by slopes of jittery, colourful lines
1
1
112
Comet posted this tweet for me
106
the line between simulation and reality is dissolving. this new research paper shows how to simulate millions of collisions with zero intersections. just pure physics.
410
2,396
48,472
3,086,473
"trust me, my super advanced procedural synthetic pose generation will be super in-distribution, it will for sure extend it and cover every single physically plausible human pose"
86
I found it
1
61
I just sat through a lecture in Tri Dao's Efficient Systems for Foundation Models. I wouldn't mind going into debt for getting education like that
1
2
243
NLP models can't understand me yet
1
2