I agree Part of working in opensource is actually transparency and being open to critique (otherwise you end up with code thieves like Ollama) I hope everyone reassess themselves and work for the good of the community
MULTIMODAL AI
-
EGO-BIRD: 100,000 Hours of Bird Footage for Autonomous Drones
By
–
after the overwhelming support for EGO-SNAKE, i’m excited to share EGO-BIRD
— Tejes Srivalsan (@tejessrivalsan) 29 mars 2026
100,000 hours of pov bird footage to train the next generation of autonomous drones pic.twitter.com/YLiNYOM444after the overwhelming support for EGO-SNAKE, i’m excited to share EGO-BIRD 100,000 hours of pov bird footage to train the next generation of autonomous drones
-
Image Models Generate Dreamlike Art Through Inpainting Techniques
By
–
Sounds like a fun rabbit hole 🙂 Using the world knowledge of image models to fill in the holes would make it particularly dream like
-
Photo albums as spatial memory palaces through AI technology
By
–
I try to over capture everything just in hopes that day will come when my photo album turns into a spatial memory palace!
-
Self-Distillation Hidden Layers Self-Supervised Vision Models
By
–
"Self-Distillation of Hidden Layers for Self-Supervised Representation Learning" This paper shows that self-supervised vision models would work much better when they predict a teacher's hidden layers across the entire visual hierarchy. As they showed that just learning from the
-
MLB Adopts Sony Computer Vision AI for Ball-Strike Calls
By
–
69% of baseball fans say they'd rather have a computer vision AI system call balls and strikes than a human umpire.
— The Rundown AI (@TheRundownAI) 29 mars 2026
This season, the MLB gave them one.
For the first time in league history, a human ump's ball-strike call is not final. A Sony computer vision system called… pic.twitter.com/MsbZYcy1ih69% of baseball fans say they'd rather have a computer vision AI system call balls and strikes than a human umpire. This season, the MLB gave them one. For the first time in league history, a human ump's ball-strike call is not final. A Sony computer vision system called
-
Meta Advances AI Object Understanding with SAM 3 and SAM 3D
By
–
SAM 3 and SAM 3D: Meta Platforms Pushes #AI Object Understanding Further
— Ronald van Loon (@Ronald_vanLoon) 29 mars 2026
by @AIatMeta#GenerativeAI #EmergingTech #Innovation #Tech #Technology pic.twitter.com/wnnZzU3Nw9SAM 3 and SAM 3D: Meta Platforms Pushes #AI Object Understanding Further
by @AIatMeta #GenerativeAI #EmergingTech #Innovation #Tech #Technology -
SWE-Vision: Teaching AI to Code Visual Intelligence
By
–
Can we unlock unprecedented visual intelligence in AI by teaching it to code what it sees? Researchers from UniPat AI and Michigan State University introduce SWE-Vision. This innovative agent allows vision-language models to write and execute Python code, using libraries like
-
Frontier Models Vision Capabilities: Benchmarks Gaming Problem
By
–
Frontier models can’t see, and if you think they can, you’ve probably been fooled by benchmarks that can totally be gamed. In the very short essay linked below I discuss a stunning new finding from Stanford that shows just how serious the problem is. And why this means a lot
-
14 most important and influential types of JEPA in AI
By
–
14 most important and influential types of JEPA ▪️ JEPA / H-JEPA
▪️ I-JEPA
▪️ MC-JEPA
▪️ V-JEPA
▪️ Audio-JEPA
▪️ Point-JEPA
▪️ 3D-JEPA
▪️ ACT-JEPA
▪️ V-JEPA 2
▪️ LeJEPA
▪️ Causal-JEPA
▪️ V-JEPA 2.1
▪️ LeWorldModel
▪️ ThinkJEPA Save the list and check this out to explore these JEPA milestones as a map of AI progress: turingpost.com/p/jepamap [Translated from EN to English]→ View original post on X — @debashis_dutta, 2026-03-29 11:51 UTC