Search papers, labs, and topics across Lattice.
University of Turin
3
0
5
Direct manipulation of latent spaces in diffusion models can lead to severe visual artifacts, but using sparse autoencoders for semantic detection offers a cleaner alternative for object erasure.
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.
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.
