Search papers, labs, and topics across Lattice.
3
0
6
The form of answer labels, not just their quantity, fundamentally shapes what LLMs learn during fine-tuning, revealing a surprising causal relationship that could redefine training strategies.
MXAttention achieves near-FP16 generation quality in video models while cutting computational costs, revolutionizing efficient attention mechanisms.
StreamKL slashes memory usage from quadratic to constant, enabling efficient long-context attention distillation on a single GPU.
