Search papers, labs, and topics across Lattice.
Samsung AI Center
4
0
5
Achieving 97.44% of optimal performance without any adapter training or internal access, ARIADNE revolutionizes how we dynamically select task-specific models at inference time.
The supposed stability of archetypal SAEs evaporates when initialization is randomized, challenging the reliability of their concept extraction claims.
LLMs are not just generating random names; they create persistent, correlated character ensembles that are infiltrating academic publishing and could undermine scholarly integrity.
Forget white-box access: this grey-box method recovers verbatim memorized content from finetuned LLMs by just comparing output logits, even revealing hidden data pipeline artifacts.
