9 citations · 17 across the 9 of their papers we have counts for
3 papers · 1 filter
Learning the identity: a case study of how SGD selects among functional decompositions
Andy Arditi, Weian Xie, David Bau +1
One might think that learning the identity function with a deep linear residual network is trivial - the path along residual connections already implements the identity, and so the…
Synthetic Persona Pretraining: Alignment from Token Zero
Julian Minder, Viktor Moskvoretskii, Raghav Singhal +12
As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical. Today, alignment, and the assistant…
Refusal in Language Models Is Mediated by a Single Direction
Andy Arditi, Oscar Obeso, Aaquib Syed +4
Conversational large language models are fine-tuned for both instruction-following and safety, resulting in models that obey benign requests but refuse harmful ones. While this ref…