#linear-attention on Hacker News

2 stories tagged linear-attention, newest first.

Screenshot of nvlabs.github.io: SANA-WM, a 2.6B open-source world model for 1-minute 720p video

SANA-WM, a 2.6B open-source world model for 1-minute 720p video (nvlabs.github.io)

384 points by mjgil 5 months ago | 145 comments

NVIDIA project page for SANA-WM, a 2.6B open-source world model that generates minute-long 720p camera-controllable video from one image on a single GPU.

world-models video-generation diffusion-transformer linear-attention

Screenshot of arxiv.org: RWKV: Reinventing RNNs for the Transformer Era

RWKV: Reinventing RNNs for the Transformer Era (arxiv.org)

318 points by ianbutler 3 years ago | 164 comments

arXiv paper introducing RWKV, a model architecture combining Transformer-style parallel training with efficient RNN inference via linear attention.

rwkv rnn transformers linear-attention

#linear-attention on Hacker News

SANA-WM, a 2.6B open-source world model for 1-minute 720p video(nvlabs.github.io)

RWKV: Reinventing RNNs for the Transformer Era(arxiv.org)

SANA-WM, a 2.6B open-source world model for 1-minute 720p video (nvlabs.github.io)

RWKV: Reinventing RNNs for the Transformer Era (arxiv.org)