Search papers, labs, and topics across Lattice.
3
0
6
7
Harnessed agentic RL can boost agent performance significantly, as shown by a 14.6-point improvement in coding tasks with minimal training data.
Text-to-video models can now learn geometrically consistent world dynamics via reinforcement learning, without expensive architectural changes.
LLM agents can slash task completion time by almost 50% simply by predicting and pre-executing likely tool calls.
