Search papers, labs, and topics across Lattice.
4
0
11
Generate consistent multi-agent videos in a shared world using a novel framework that fuses multi-view data and cross-agent attention.
LLMs respond to increasingly difficult out-of-distribution inputs by activating sparser representations in their last hidden states, revealing a quantifiable relationship between task difficulty and neural activity.
Stop hand-crafting RLHF curricula: ACTOR-CURATOR learns to dynamically select training problems, boosting performance by up to 30% and speeding up training by 80% on challenging reasoning tasks.
Agentic AI can automate complex optical systems control with near-perfect success rates, leaving code-generation approaches in the dust.
