TECH·22 hours agoDeconstructing the Kimi K3 Technical Report: Moonshot AI Starts Setting Problems for DeepSeek as Two Scaling Axes Redefine the FrontierPandaily
PAPER·5 days agoSANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video GenerationHugging Face Papers
TECH·June 1, 2026Parallax: A Parameterized Local Linear Attention That Keeps Softmax and Adds a Learned Covariance Correction BranchMarkTechPost
PAPER·May 29, 2026Parallax: Parameterized Local Linear Attention for Language ModelingHugging Face Papers
PAPER·May 26, 2026HorizonStream: Long-Horizon Attention for Streaming 3D ReconstructionHugging Face Papers
TECH·May 24, 2026NVIDIA AI Releases Gated DeltaNet-2: A Linear Attention Layer That Decouples Erase and Write in the Delta RuleMarkTechPost
PAPER·May 22, 2026Gated DeltaNet-2: Decoupling Erase and Write in Linear AttentionHugging Face Papers
TECH·May 18, 2026NVIDIA Announces SANA-WM, an AI That Can Generate 1-Minute Videos and Precisely Control Camera MovementGIGAZINE
PAPER·May 15, 2026SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion TransformerHugging Face Papers
PAPER·May 13, 2026Lite3R: A Model-Agnostic Framework for Efficient Feed-Forward 3D ReconstructionHugging Face Papers
