Search papers, labs, and topics across Lattice.
3
0
7
FreeAct boosts quantized LLM performance by dynamically adapting activation transformations to different token types, moving beyond the static transformations that limit existing methods.
LLMs and LVLMs share more than half their top-activated neurons during multi-step inference, opening a surprisingly cheap path to boost vision-language reasoning by transplanting skills from text-only models.
Achieve SOTA joint audio-video generation with JavisDiT++ using just 1M public training examples, rivaling performance of models trained on proprietary datasets.
