Refusal in language models is mediated by a single direction (arxiv.org) | HN Screenshots

YHacker Newsscreenshots

Screenshot of arxiv.org: Refusal in language models is mediated by a single direction

First screen, captured 2 years ago.

Looks similar

Screenshot of arxiv.org: Refusal in Language Models Is Mediated by a Single Direction

Refusal in Language Models Is Mediated by a Single Direction (arxiv.org)

101 points by fagnerbrack 5 months ago | 36 comments

Screenshot of arxiv.org: Recursive Language Models

Recursive Language Models (arxiv.org)

135 points by schmuhblaster 9 months ago | 23 comments

Screenshot of arxiv.org: An Empirical Study of Mamba-Based Language Models

An Empirical Study of Mamba-Based Language Models (arxiv.org)

39 points by panabee 2 years ago | 2 comments

Screenshot of arxiv.org: I'm afraid I can't do that: Prompt refusal in generative language models

I'm afraid I can't do that: Prompt refusal in generative language models (arxiv.org)

155 points by belter 3 years ago | 138 comments

Screenshot of arxiv.org: AnyMAL: An Efficient and Scalable Any-Modality Augmented Language Model

AnyMAL: An Efficient and Scalable Any-Modality Augmented Language Model (arxiv.org)

26 points by babakd 3 years ago | 4 comments

Screenshot of arxiv.org: Unsupervised Elicitation of Language Models

Unsupervised Elicitation of Language Models (arxiv.org)

128 points by kordlessagain 1 year ago | 18 comments