Everyone assumed semantic encoders couldn’t reconstruct images.
Turns out, they can and better than VAEs. RAE + DINOv2-B achieves 0.49 rFID vs 0.62 for SD-VAE, with 6× less compute. That’s fewer FLOPs and sharper reconstructions.
RESEARCH
-

RAE + DINOv2-B beats VAEs with sharper images
By
–
-

RAEs improve diffusion models over VAEs
By
–
Today, most diffusion models still use VAEs built on 2021 tech. They compress images into low-dimensional latents (like 4 channels). That’s why diffusion models lose global structure and texture fidelity. RAEs fix this by encoding rich semantic features directly from pretrained
-

New Paper Revolutionizes Diffusion Models
By
–
Holy shit…Diffusion just leveled up A new paper “Diffusion Transformers with Representation Autoencoders” basically kills the VAE era. Instead of the old VAE bottleneck, they use representation autoencoders (RAEs) built from pretrained encoders like DINO or SigLIP. The
-

Open datasets trend enables independent model training
By
–
I love the diversity of trending open datasets these days. There’s no excuse anymore not to train your own models! – Fineweb and a shuffle of it by @karpathy – Webscale-RL, a large-scale reinforcement learning dataset from @salesforce – SVQ, an audio dataset from @Google –
-
Top 5 Technology Trends Reshaping 2026 Innovation
By
–
The Top 5 Technology Trends for 2026 From intelligent agents and space computing to bio-tech and digital twins, these are the five technological shifts set to reshape business, society and innovation in 2026. Read more https://
bernardmarr.com/the-top-5-tech
nology-trends-for-2026/
… #TechTrends #Innovation -
Elon Musk’s Distributed AI Approach to AGI Development
By
–
I think early on @elonmusk was talking about how AGI could come from millions of distributed AIs instead of one central system
-

AGI Definition: Cognitive Versatility Beyond Single Task Excellence
By
–
Finally, we have a definition for AGI! A recent paper breaks down what it actually means to reach human-level AI, and the results are surprising. Here's what matters: AGI isn't about passing a single test or excelling at one task. It's about matching the cognitive versatility
-

AI’s Future Challenge: Efficiency Over Intelligence and Power Demands
By
–
The true test of AI’s future won’t be intelligence, it will be efficiency. As AI power demand accelerates toward 1,600 TWh by 2035, we’re approaching a crossroads: innovation is outpacing infrastructure. The same technology built to optimize global systems now risks overloading
-
AI’s Role in Global Energy Transformation at ADIPEC 2025
By
–
What happens when artificial intelligence becomes the engine of global energy transformation?
— Harold Sinnott #MWC26 (@HaroldSinnott) 18 octobre 2025
At #ADIPEC2025, we’ll find out, and shape what’s next.
Under the theme "Energy. Intelligence. Impact.", ADIPEC 2025 will explore how digitalization, climate tech, and cross-sector… pic.twitter.com/Y3hS3ssSLVWhat happens when artificial intelligence becomes the engine of global energy transformation? At #ADIPEC2025, we’ll find out, and shape what’s next. Under the theme "Energy. Intelligence. Impact.", ADIPEC 2025 will explore how digitalization, climate tech, and cross-sector