Thomas Kipf
@tkipf
Followers
28K
Following
11K
Media
260
Statuses
2K
Research at @GoogleDeepMind. Controllable World Simulators (GNNs, Structured World Models, Neural Assets). Veo Controls (Ingredients & more).
San Francisco, CA
Joined June 2009
My PhD thesis "Deep Learning with Graph-Structured Representations" is now available for download: -- It covers a range of emerging topics in Deep Learning: from graph neural nets (and graph convolutions) to structure discovery (objects, relations, events)
42
611
3K
Regularly check in with your loved ones and friends even if (or especially if) you live in different places — things can take a dark turn very quickly, unfortunately. Devastating news….
Heartbroken to see another suicide note circulating in AI twitter. Work is never more important than life. Please take care of yourself and those around you.
0
2
23
RT @docmilanfar: Career "advice":. Live modestly, save money.Put your family and health first.Keep learning - don't live on your laurels.No….
0
21
0
Bullish on world models. Congrats to the team - incredible progress! 🚀.
What if you could not only watch a generated video, but explore it too? 🌐. Genie 3 is our groundbreaking world model that creates interactive, playable environments from a single text prompt. From photorealistic landscapes to fantasy realms, the possibilities are endless. 🧵
1
1
26
Gemini 2.5 Deep Think is out now for Google AI Ultra subscribers!
We’re bringing a version of Deep Think that achieved gold-medal status at IMO to Ultra subscribers in the @Geminiapp (+ the official version is now in the hands of mathematicians). Toggle it on when reasoning through complex scientific literature, tackling a coding problem that
1
0
15
Video models (here: Veo) are more general reasoners than you’d think.
We just discovered the 🔥 COOLEST 🔥 trick in Flow that we have to share:. Instead of wordsmithing the perfect prompt, you can just. draw it. Take the image of your scene, doodle what you'd like on it (through any editing app), and then briefly describe what needs to happen
0
0
22
RT @DynamicWebPaige: 💖 Loved this @lexfridman conversation with Demis!. His excitement about how our video model, Veo 3, is learning intuit….
0
4
0
RT @shaneguML: "Prioritization". I prioritized what to do based on a simple metric "(impact remaining)/(# top talent working)". Each resear….
0
4
0
1) Intelligence is the process of minimizing the generator-verifier gap.2) The generator-verifier gap is a fundamental property of the universe.3) ASI is the optimal such process.
If a solution is fundamentally easier to be verified than to be generated, it probably means that there is a learning signal. Lots of problems fall in this category.
5
3
75
Open position in the Veo team - come join us!.
Want to be part of a team redefining SOTA for generative video models? Excited about building models that can reach billions of users? The Veo team is hiring! We are looking for amazing researchers and engineers, in North America and Europe. Details below:
1
0
15
RT @AndrewLampinen: For humans, mathematical symbols (and formal systems like lean) are *tools* we learn how to use, not a structure that w….
0
11
0
Official gold-medal performance with Gemini at IMO. Massive congrats to the team!. “This year, we were amongst an inaugural cohort to have our model results officially graded and certified by IMO coordinators using the same criteria as for student solutions.”.
Advanced version of Gemini Deep Think (announced at #GoogleIO) using parallel inference time computation achieved gold-medal performance at IMO, solving 5/6 problems with rigorous proofs as verified by official IMO judges! Congrats to all involved!.
1
2
31
Look at that gap.
We just released the evaluation of LLMs on the 2025 IMO on MathArena! Gemini scores best, but is still unlikely to achieve the bronze medal with its 31% score (13/42). 🧵(1/4)
0
0
8
Great initiative 👇.
NeurIPS is pleased to officially endorse EurIPS, an independently-organized meeting taking place in Copenhagen this year, which will offer researchers an opportunity to additionally present their accepted NeurIPS work in Europe, concurrently with NeurIPS. Read more in our blog.
0
0
10
RT @LogConference: We’re thrilled to share that the first in-person LoG conference is officially happening December 10–12, 2025 at Arizona….
logconference.org
Learning on Graphs Conference
0
26
0
I use them daily since a couple of months and they're amazing. The AI features are hit-and-miss, though. Photos/videos & audio/mic are awesome. Touch pad activates too easily.
After wearing Ray-Ban Meta Wayfarer glasses for a few weeks . I feel kind of naked wearing regular sunglasses. I've found three use-cases that are hard to roll back:. * Spontaneous photos of my kids when we're out and about. Any cool pose that has a half-life of 3 seconds I.
0
0
2