> Together with SpaceXAI, we're training a significantly larger model from scratch, using 10x more total compute. With Colossus 2's million H100-equivalents and our combined data and training techniques, we expect this to be a major leap in model capability. That's double
RESEARCH
-
Composer 2.5 and Large-Scale AI Model Training Initiative
By
–
Composer 2.5 is built on top of Kimi K2.5 Also interesting > Together with SpaceXAI, we're training a significantly larger model from scratch, using 10x more total compute. With Colossus 2's million H100-equivalents and our combined data and training techniques, we expect
-
New Large-Scale AI Model Training with Increased Compute
By
–
Together with SpaceXAI, we’re training a significantly larger model from scratch, using 10x more total compute. With Colossus 2’s million H100-equivalents and our combined data and training techniques, we expect this to be a major leap in model capability.
-
Technical Improvements to Composer AI Training and RL Methods
By
–
We improved Composer by scaling training, generating more complex RL environments, and introducing new learning methods. For example, we use text feedback during RL to learn faster by assigning credit in rollouts spanning hundreds of thousands of tokens.
-
New AI tool could replace costly cancer gene expression profiling
By
–
New AI tool could replace costly cancer gene expression profiling. https://
medicalxpress.com/news/2026-05-a
i-tool-cancer-gene-profiling.html
… #AI #Cancer #Healthcare #research -
Human preferences vs model learning
By
–
Humans can invent preferences faster than models can learn to satisfy them
-
New Research Introduces Adversarial Parameter Decomposition for Neural Network Interpretability
By
–
The "model weights are unreadable" excuse just died in one paper.
— AlphaSignal AI (@AlphaSignalAI) 18 mai 2026
Neural networks have billions of parameters.
Nobody really knows what each one does.
A new paper introduces adVersarial Parameter Decomposition.
The method splits a model's weights into small,… pic.twitter.com/izUeHRCQ5fThe "model weights are unreadable" excuse just died in one paper. Neural networks have billions of parameters. Nobody really knows what each one does. A new paper introduces adVersarial Parameter Decomposition. The method splits a model's weights into small,
-
A Closer Look Into The Math Behind Neural Networks
By
–
A Closer Look Into The Math Behind Neural Networks! #BigData #Analytics #DataScience #AI #MachineLearning #IoT #IIoT #PyTorch #Python #RStats #TensorFlow #GoLang #CloudComputing #Serverless #DataScientist #Linux #Programming #Coding #100DaysofCode https://
geni.us/Math-Neural-Ne
tworks
… -
Models show consistent theory-of-mind failures
By
–
Its a consistent theory-of-mind failure in models that are otherwise suprisingly good at theory-of-mind
-
Unified Vision World Models framework from Beijing Jiaotong, ByteDance, Tencent
By
–
What if AI could learn the world just by watching? Researchers from Beijing Jiaotong, ByteDance, Tencent present a unified framework for Vision World Models: encoding visuals, learning dynamics, simulating outcomes. This survey outperforms fragmented taxonomies, outlining