Great collab with @SakanaAILabs on an #ICML26 paper about sparse transformer kernels + formats optimized for modern NVIDIA GPU execution. • TwELL sparse packing
• Fused CUDA kernels
• 20%+ inference/training speedups at scale Paper + code below
RESEARCH
-
New ICML26 Paper on Sparse Transformer Kernels for NVIDIA GPUs
By
–
-
Representational Geometry Influences Uncertainty in Large Language Models
By
–
"Representational Curvature Modulates Behavioral Uncertainty in LLMs" Most interpretability work studies what features live inside LLMs. But this paper studies something deeper, which is the geometry of the whole evolving representation. The key result is that representational
-
Optimiser les LLM avec la sparsité adaptée au GPU
By
–
How do we make LLMs faster and lighter? Don’t force the GPU to adapt to sparsity. Reshape the sparsity to fit the GPU! ⚡️
— Sakana AI (@SakanaAILabs) 8 mai 2026
Excited to share our new #ICML2026 paper in collaboration with @NVIDIA: "Sparser, Faster, Lighter Transformer Language Models". This work introduces new… pic.twitter.com/ehByWHIh6IHow do we make LLMs faster and lighter? Don’t force the GPU to adapt to sparsity. Reshape the sparsity to fit the GPU! Excited to share our new #ICML2026 paper in collaboration with @NVIDIA
: "Sparser, Faster, Lighter Transformer Language Models". This work introduces new -
Continuous-Time Distribution Matching for Few-Step Diffusion Distillation
By
–
Continuous-Time Distribution Matching for Few-Step Diffusion Distillation paper: https://
huggingface.co/papers/2605.06
376
… -
Apple introduces TIDE: Every Layer Knows the Token Beneath the Context
By
–
Apple presents TIDE Every Layer Knows the Token Beneath the Context paper: https://
huggingface.co/papers/2605.06
216
… -
LLMs’ Math Progress Accelerates Rapidly
By
–
It's surprising to see the pace of progress of LLMs (which until not long ago were very bad at math) achieving ever greater milestones. Take a look at the graph above. The progress curve is getting faster and faster. I'll leave the link to the paper here: https://
arxiv.org/abs/2605.06651 -
MARBLE: Multi-Aspect Reward Balance for Diffusion RL
By
–
MARBLE Multi-Aspect Reward Balance for Diffusion RL paper: https://
huggingface.co/papers/2605.06
507
… -
AI for Mental Health Symposium at Stanford
By
–
As part of the @Stanford Health AI Week, join us for the inaugural AI for Mental Health (AI4MH) Symposium on June 1. The event will explore how AI is shaping mental health – from foundational advances to real-world implementation and responsible use. https://
med.stanford.edu/psychiatry/spe
cial-initiatives/ai4mh/events/symposium2026.html
… -
Google reclaims lead in FrontierMath Tier 4
By
–
LEADER in FRONTIER MATH T4! New record in one of the most challenging math benchmarks -FrontierMath Tier 4- where Google has just reclaimed the lead with a 47.9% !!, stealing the position from GPT 5.5 Pro It does it with its new math agent
-
New Continuous Latent Diffusion Language Model paper
By
–
Continuous Latent Diffusion Language Model paper: https://
huggingface.co/papers/2605.06
548
…