Search papers, labs, and topics across Lattice.
2
4
4
7
Omni-modal foundation models achieve self-improvement via RL compute scaled to 1M context, leveraging asynchronous training and agentic grading for enhanced intelligence.
RL gains in LLMs are surprisingly localized to a low-dimensional, non-infinitely compressible manifold, with control directions orthogonal to principal activation components.
