24.
FlashAR claims 22x faster autoregressive image generation
A lightweight vertical prediction head enables parallel decoding for existing autoregressive image models without retraining from scratch
@jiqizhixin on Backlist
3 appearances on the backlist front page in the last 30 days.
51.
There is now a smarter way to pick data for training LLMs!
There is now a smarter way to pick data for training LLMs! Enter OPUS! This is an ICML Oral paper from SJTU, Alibaba, UW–Madison, UIUC, and Mila - Quebec AI Institute. The proposed method dynamically and intelligently selects the most im
46.
New from NVIDIA!
New from NVIDIA! You can edit a model’s compressed memory without scrambling what it already knows! Enter Gated DeltaNet-2. It separates the erase and write operations in linear attention using two independent gates – one for forgetting