We have 3 of 56 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
An expert-driven methodology for constructing such alignment data and applying it to a normative framework grounded in Islamic ethical, theological, and jurisprudential traditions is presented.
This work uncovers a structured internal representation of refusal in large language models: a reusable core of shared refusal latents supplemented by style- and domain-specific latents, and clarifies the limits of linear interpretability for aligned model behavior.
A comparative study of three methods that identify language-controlling latents in cross-layer transcoders: activation value-based selection (ValSel), activation frequency-based selection (FreqSel), and LLM-generated latent annotation-based selection (AnnSel) finds all three effectively manipulate generation language.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
