Search papers, labs, and topics across Lattice.
2
0
5
7
MLLMs can "think" with images, but their actions often don't match their reasoning, and this paper solves that with a new training method that forces them to explain what they see.
FreeAct boosts quantized LLM performance by dynamically adapting activation transformations to different token types, moving beyond the static transformations that limit existing methods.
