Search papers, labs, and topics across Lattice.
3
0
3
EVON transforms structured weight uncertainty into a practical tool, yielding superior performance in language model pretraining without the overhead of complex implementations.
Data-similarity can reliably guide output tracing, but its accuracy skyrockets when refined with data-influence, revealing a powerful synergy between the two measures.
Merging past knowledge with fast adaptation in the CoVON optimizer leads to superior performance in continual learning tasks, outperforming traditional methods.
