Search papers, labs, and topics across Lattice.
3
0
5
Curation of training data for brevity can yield a staggering 35x improvement in inference efficiency without sacrificing accuracy in VLMs.
Cut KV-cache transfer times by up to 32% with SplitZip, a new GPU-friendly lossless compressor that unlocks faster disaggregated LLM serving.
Stop wasting your finetuning data: Specialized Pretraining (SPT) can outperform standard pretraining and finetuning, achieving better domain performance with fewer parameters and less compute.
