Marius Slavescu (@GTARobotics) profile picture

STEMCA Inventor co-author, robotics/STEM education and DIY/maker platform, Founder of GTA Robotics community and Open Source Self Driving Car project https://teapawt.net/t.co/xcxLnEZKjN

Richmond Hill, Ontario
Joined April 2016
Photo by Marius Slavescu (@GTARobotics)Photo by Marius Slavescu (@GTARobotics)Photo by Marius Slavescu (@GTARobotics)Photo by Marius Slavescu (@GTARobotics)
Replying to @AndrewYNg
@AndrewYNg We can do it by enabling everyone to collect, analyze, share data and code! I try this in #OpenSource #SelfDrivingCar Initiative teapawt.net/AndrewYNg/status/82834…
For AI to be free we need not just Open Source, but also a strong Open Data movement.
1
1
19
Impressive work towards enhancing #robotics manipulations! Congratulation @andreslavescu and @physicreality team!
[1/2] “The Task Selects the Hand” Today we introduce a method for synthesizing families of gripper morphologies using reinforcement learning. Starting from a single first-person video of an everyday task, we extract the hand, object, and contact trace; reconstruct the scene into a physics-ready simulation environment; and compose and train a gripper for that task, rather than forcing the task onto an arbitrary gripper. Every candidate that succeeds in simulation is then subjected to static finite element analysis under the working load, and any hand that cannot carry it is vetoed. The method closes with a printable CAD file describing the resulting gripper. The four finger hand produced in this blog, PR-HAND-F4J2, did not exist before the video. Link below 🧵
1
183
[1/2] In collaboration with @ContractedAI: "Knowing, but not Noticing" - We show that frontier ER models (which robot foundation models dominantly initialize from) consistently fail to know 'what to do next' in a chained task; - how to measure that failure quantitatively using data sourced from a novel branching/ 'counterfactual'-style ego data collection regime; and - propose a means to produce post-training data to fill in those gaps, based on what context ablations via harnessing reveal the model already knows, but simply fails to conjure reflexively on its own. Link in thread 🧵
3
8
3
36
91,772
Marius Slavescu retweeted
The ability to try any new thing with a sensible approach is a cornerstone of artificial general intelligence. Now the real world is yours for learning, NEO.
18
28
8
347
23,817
Marius Slavescu retweeted
mHC.cu performance oriented changes to enable larger expansion rates (n): 1. Template specialization for n=4/8/16/32. Kernels now dispatch to optimal unroll # instead of always using 32 2. Tensor Core path for n >= 32 via cuBLASLt - restructured to row-major layout so no transpose kernels needed 3. enabling pipelining for larger expansion rates (n) for the static H path. pipelining active at n >= 16.
6
4
58
3,971
Marius Slavescu retweeted
Wrote a fused mHC layer in pytorch using the kernels as a backend, and achieved some interesting speedups! Tested on H100 SXM5
Decided to implement Deepseek's new paper (mHC: Manifold-Constrained Hyper-Connections) in CUDA.
18
39
3
561
48,609
Marius Slavescu retweeted
Decided to implement Deepseek's new paper (mHC: Manifold-Constrained Hyper-Connections) in CUDA.
29
36
9
970
179,501
Marius Slavescu retweeted
Update: INT8 path now has fused perturbed matmul for better performance. MLP expand uses pure INT8 activations with no float conversion: github.com/AndreSlavescu/egg… per-step time on H100 PCI-e is at <60ms. @j_foerst @bidiptas13 @d0rc
1
1
11
488
Marius Slavescu retweeted
EGGROLL 8xB200 distributed training with P2P weight averaging! github.com/AndreSlavescu/egg… Thank you @NorthflankWill and @northflank for the B200s.
4
7
1
40
2,049
It's @ArtificialAnlys official: Parakeet TDT running on @TeamHathora has the best cost to WER (word error rate) ratio of any ASR model! This was all in less than a month since we launched artificialanalysis.ai/speech… Huge thanks to the @nvidia voice research team (Adi & folks) and the AA team (@grmcameron @kiriillbutler) Try it for yourself here: models.hathora.dev/model/nvi…
3
1
8
379
Marius Slavescu retweeted
Tonight’s AI Lightning Talks at the @shinecapital office with talks from Zack @AssemblyAI @andre_slav03 @TeamHathora And Kent from @Medal_TV and General Intuition. We lit up the minds of NYC AI curious. Come back in 2026!
2
3
296
Super cool viz of how an Omni modal model can process audio tokens in/out by @andre_slav03 qwen3-omni-mechanics-gui-web…
2
8
200
Marius Slavescu retweeted
Here are our three papers on ways to use AI for teaching. All have prompts & details: Using AI to make teaching easier & more impactful: papers.ssrn.com/sol3/papers.… Using errors made by AI to teach hard things: papers.ssrn.com/sol3/papers.… Assigning AI to students: papers.ssrn.com/sol3/papers.…
26
247
22
1,015
298,032
Thank you @Autodromo_Monza for the completion certificate of the first race on a non-oval circuit during MIMO 2023 🏁🤖🏎️ #MilanoMonza #IndyAutonomousChallenge #NoSteeringWheelNoProblem #AutonomousRacing #MIMO2023
1
3
6
620
“More than 10,000 spectators filled the stands to witness the driverless racecars competing on the Monza “Temple of Speed” Formula One circuit as part of the Milan Monza Motor Show (MIMO) that ran last weekend at the @Autodromo_Monza” - @IAutonomousV insideautonomousvehicles.com…
4
8
1,195
We’re showing @AndrettiFE’s @JakeDennis19 around Indy today. During our first stop, he’s seeing the innovations in autonomous mobility happening right here in Indiana with @IndyAChallenge. – at Indianapolis, IN
9
19
2,363
Congratulations to autonomous racing team MITT-PITT-RW who achieved a personal best time of 173.8 kph on their first @IndyAChallenge road race in Monza, Italy 🤖🏎️ The first challenge? Learning to turn right! Full Story: bit.ly/3XoP2GG | #UWaterlooNews
1
8
1
15
5,649
Marius Slavescu retweeted
Congratulations to the winners and finalists of the Best Paper Award of the 4th Workshop on Event-based Vision at #CVPR2023: Papers and code: tub-rip.github.io/eventvisio… #eventcamera With @guillermogb @KostasPenn @CFermuller @DMigliorePro
1
7
16
2,706
Marius Slavescu retweeted
Carve out a few hours to learn streamlit.io/ Powerful for rapid prototyping, interactive visualization. It's a hammer and you'll start seeing a lot of nails.
105
338
51
3,531
567,290