Search papers, labs, and topics across Lattice.
3
11
6
6
SKIM compresses procedural skills in LLMs by 30-60% without sacrificing performance, revolutionizing how we manage reusable natural language skills.
Explicitly enumerating skills in-context doesn't scale for agentic LLMs, but retrieving skills on demand can substantially improve performance – if the LLM can figure out when and which skill to load.
LLMs still struggle to learn effectively from user feedback during service, as revealed by a new benchmark spanning multiple domains and languages.
