Search papers, labs, and topics across Lattice.
2
0
4
3
Mechanistic interpretability gets a formal footing: "Certified Circuits" uses data subsampling to find provably stable sub-networks, boosting accuracy by up to 91% while using 45% fewer neurons.
LLMs can now perform structured expert risk elicitation in minutes, rivaling human expert panels and slashing the months-long timelines of traditional Delphi methods.
