New Research at ICML 2026: Graph Models and Diffusion
4h agoen
From the article
Key works in graph foundation models and diffusion transformer optimization were presented at the ICML 2026 conference. The industry is shifting from specialized GNNs to universal graph foundation models and seeking ways to scale diffusion efficiently.
Continue reading on mlllm.ioYou might also wanna read
How Open Models Are Driving AI Research
NVIDIA Developer Forums·3d ago
How Open Models Are Driving AI Research
NVIDIA Developer Forums·3d ago
How Open Models Are Driving AI Research
NVIDIA Developer Forums·3d ago
[ICML 2026] The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models
arxiviq.substack.com·3d ago
Google's Presence at ICML 2026: Diamond Sponsorship and Research Contributions
Google is a Diamond Sponsor of ICML 2026, the 43rd International Conference on Machine Learning, held in Seoul, South Korea from July 6–11,
goo.gle·2d agoTogether AI at ICML 2026: frontier research across the full stack
Together·9d ago
[ICML 2026] High-accuracy sampling for diffusion models and log-concave distributions
arxiviq.substack.com·2d ago
Tauformer: A Topological Transformer Architecture Using Laplacian-Derived Scalar Attention
The article discusses Tauformer, a novel topological transformer architecture that replaces traditional dot-product attention with a Laplaci
tuned.org.uk·5mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.